Gilberto Ochoa-Ruiz

Mansoor Ali, Gilberto Ochoa-Ruiz, Sharib Ali

Surgical tool detection in minimally invasive surgery is an essential part of computer-assisted interventions. Current approaches are mostly based on supervised methods which require large fully labeled data to train supervised models and suffer from pseudo label bias because of class imbalance issues. However large image datasets with bounding box annotations are often scarcely available. Semi-supervised learning (SSL) has recently emerged as a means for training large models using only a modest amount of annotated data; apart from reducing the annotation cost. SSL has also shown promise to produce models that are more robust and generalizable. Therefore, in this paper we introduce a semi-supervised learning (SSL) framework in surgical tool detection paradigm which aims to mitigate the scarcity of training data and the data imbalance through a knowledge distillation approach. In the proposed work, we train a model with labeled data which initialises the Teacher-Student joint learning, where the Student is trained on Teacher-generated pseudo labels from unlabeled data. We propose a multi-class distance with a margin based classification loss function in the region-of-interest head of the detector to effectively segregate foreground classes from background region. Our results on m2cai16-tool-locations dataset indicate the superiority of our approach on different supervised data settings (1%, 2%, 5%, 10% of annotated data) where our model achieves overall improvements of 8%, 12% and 27% in mAP (on 1% labeled data) over the state-of-the-art SSL methods and a fully supervised baseline, respectively. The code is available at https://github.com/Mansoor-at/Semi-supervised-surgical-tool-det

David Laines, Gissella Bejarano, Miguel Gonzalez-Mendoza et al.

Sign Language Recognition (SLR) systems aim to be embedded in video stream platforms to recognize the sign performed in front of a camera. SLR research has taken advantage of recent advances in pose estimation models to use skeleton sequences estimated from videos instead of RGB information to predict signs. This approach can make HAR-related tasks less complex and more robust to diverse backgrounds, lightning conditions, and physical appearances. In this work, we explore the use of a spatio-temporal skeleton representation such as Tree Structure Skeleton Image (TSSI) as an alternative input to improve the accuracy of skeleton-based models for SLR. TSSI converts a skeleton sequence into an RGB image where the columns represent the joints of the skeleton in a depth-first tree traversal order, the rows represent the temporal evolution of the joints, and the three channels represent the (x, y, z) coordinates of the joints. We trained a DenseNet-121 using this type of input and compared it with other skeleton-based deep learning methods using a large-scale American Sign Language (ASL) dataset, WLASL. Our model (SL-TSSI-DenseNet) overcomes the state-of-the-art of other skeleton-based models. Moreover, when including data augmentation our proposal achieves better results than both skeleton-based and RGB-based models. We evaluated the effectiveness of our model on the Ankara University Turkish Sign Language (TSL) dataset, AUTSL, and a Mexican Sign Language (LSM) dataset. On the AUTSL dataset, the model achieves similar results to the state-of-the-art of other skeleton-based models. On the LSM dataset, the model achieves higher results than the baseline. Code has been made available at: https://github.com/davidlainesv/SL-TSSI-DenseNet.

Francisco Lopez-Tiro, Juan Pablo Betancur-Rengifo, Arturo Ruiz-Sanchez et al.

Knowing the cause of kidney stone formation is crucial to establish treatments that prevent recurrence. There are currently different approaches for determining the kidney stone type. However, the reference ex-vivo identification procedure can take up to several weeks, while an in-vivo visual recognition requires highly trained specialists. Machine learning models have been developed to provide urologists with an automated classification of kidney stones during an ureteroscopy; however, there is a general lack in terms of quality of the training data and methods. In this work, a two-step transfer learning approach is used to train the kidney stone classifier. The proposed approach transfers knowledge learned on a set of images of kidney stones acquired with a CCD camera (ex-vivo dataset) to a final model that classifies images from endoscopic images (ex-vivo dataset). The results show that learning features from different domains with similar information helps to improve the performance of a model that performs classification in real conditions (for instance, uncontrolled lighting conditions and blur). Finally, in comparison to models that are trained from scratch or by initializing ImageNet weights, the obtained results suggest that the two-step approach extracts features improving the identification of kidney stones in endoscopic images.

Axel Garcia-Vega, Ricardo Espinosa, Luis Ramirez-Guzman et al.

Endoscopy is the most widely used imaging technique for the diagnosis of cancerous lesions in hollow organs. However, endoscopic images are often affected by illumination artefacts: image parts may be over- or underexposed according to the light source pose and the tissue orientation. These artifacts have a strong negative impact on the performance of computer vision or AI-based diagnosis tools. Although endoscopic image enhancement methods are greatly required, little effort has been devoted to over- and under-exposition enhancement in real-time. This contribution presents an extension to the objective function of LMSPEC, a method originally introduced to enhance images from natural scenes. It is used here for the exposure correction in endoscopic imaging and the preservation of structural information. To the best of our knowledge, this contribution is the first one that addresses the enhancement of endoscopic images using deep learning (DL) methods. Tested on the Endo4IE dataset, the proposed implementation has yielded a significant improvement over LMSPEC reaching a SSIM increase of 4.40% and 4.21% for over- and underexposed images, respectively.

Axel Garcia-Vega, Ricardo Espinosa, Gilberto Ochoa-Ruiz et al.

Endoscopy is the most widely used medical technique for cancer and polyp detection inside hollow organs. However, images acquired by an endoscope are frequently affected by illumination artefacts due to the enlightenment source orientation. There exist two major issues when the endoscope's light source pose suddenly changes: overexposed and underexposed tissue areas are produced. These two scenarios can result in misdiagnosis due to the lack of information in the affected zones or hamper the performance of various computer vision methods (e.g., SLAM, structure from motion, optical flow) used during the non invasive examination. The aim of this work is two-fold: i) to introduce a new synthetically generated data-set generated by a generative adversarial techniques and ii) and to explore both shallow based and deep learning-based image-enhancement methods in overexposed and underexposed lighting conditions. Best quantitative results (i.e., metric based results), were obtained by the deep-learnnig-based LMSPEC method,besides a running time around 7.6 fps)

Elias Villalvazo-Avila, Francisco Lopez-Tiro, Jonathan El-Beze et al.

This contribution presents a deep learning method for the extraction and fusion of information relating to kidney stone fragments acquired from different viewpoints of the endoscope. Surface and section fragment images are jointly used during the training of the classifier to improve the discrimination power of the features by adding attention layers at the end of each convolutional block. This approach is specifically designed to mimic the morpho-constitutional analysis performed in ex-vivo by biologists to visually identify kidney stones by inspecting both views. The addition of attention mechanisms to the backbone improved the results of single view extraction backbones by 4% on average. Moreover, in comparison to the state-of-the-art, the fusion of the deep features improved the overall results up to 11% in terms of kidney stone classification accuracy.

Rafael Martinez-Garcia-Peña, Mansoor Ali Teevno, Gilberto Ochoa-Ruiz et al.

Domain shift is a well-known problem in the medical imaging community. In particular, for endoscopic image analysis where the data can have different modalities the performance of deep learning (DL) methods gets adversely affected. In other words, methods developed on one modality cannot be used for a different modality. However, in real clinical settings, endoscopists switch between modalities for better mucosal visualisation. In this paper, we explore the domain generalisation technique to enable DL methods to be used in such scenarios. To this extend, we propose to use super pixels generated with Simple Linear Iterative Clustering (SLIC) which we refer to as "SUPRA" for SUPeRpixel Augmented method. SUPRA first generates a preliminary segmentation mask making use of our new loss "SLICLoss" that encourages both an accurate and color-consistent segmentation. We demonstrate that SLICLoss when combined with Binary Cross Entropy loss (BCE) can improve the model's generalisability with data that presents significant domain shift. We validate this novel compound loss on a vanilla U-Net using the EndoUDA dataset, which contains images for Barret's Esophagus and polyps from two modalities. We show that our method yields an improvement of nearly 20% in the target domain set compared to the baseline.

Daniel Flores-Araiza, Francisco Lopez-Tiro, Jonathan El-Beze et al.

Identifying the type of kidney stones can allow urologists to determine their cause of formation, improving the prescription of appropriate treatments to diminish future relapses. Currently, the associated ex-vivo diagnosis (known as Morpho-constitutional Analysis, MCA) is time-consuming, expensive and requires a great deal of experience, as it requires a visual analysis component that is highly operator dependant. Recently, machine learning methods have been developed for in-vivo endoscopic stone recognition. Deep Learning (DL) based methods outperform non-DL methods in terms of accuracy but lack explainability. Despite this trade-off, when it comes to making high-stakes decisions, it's important to prioritize understandable Computer-Aided Diagnosis (CADx) that suggests a course of action based on reasonable evidence, rather than a model prescribing a course of action. In this proposal, we learn Prototypical Parts (PPs) per kidney stone subtype, which are used by the DL model to generate an output classification. Using PPs in the classification task enables case-based reasoning explanations for such output, thus making the model interpretable. In addition, we modify global visual characteristics to describe their relevance to the PPs and the sensitivity of our model's performance. With this, we provide explanations with additional information at the sample, class and model levels in contrast to previous works. Although our implementation's average accuracy is lower than state-of-the-art (SOTA) non-interpretable DL models by 1.5 %, our models perform 2.8% better on perturbed images with a lower standard deviation, without adversarial training. Thus, Learning PPs has the potential to create more robust DL models.

Pedro E. Chavarrias-Solanon, Mansoor Ali-Teevno, Gilberto Ochoa-Ruiz et al.

Prevalence of gastrointestinal (GI) cancer is growing alarmingly every year leading to a substantial increase in the mortality rate. Endoscopic detection is providing crucial diagnostic support, however, subtle lesions in upper and lower GI are quite hard to detect and cause considerable missed detection. In this work, we leverage deep learning to develop a framework to improve the localization of difficult to detect lesions and minimize the missed detection rate. We propose an end to end student-teacher learning setup where class probabilities of a trained teacher model on one class with larger dataset are used to penalize multi-class student network. Our model achieves higher performance in terms of mean average precision (mAP) on both endoscopic disease detection (EDD2020) challenge and Kvasir-SEG datasets. Additionally, we show that using such learning paradigm, our model is generalizable to unseen test set giving higher APs for clinically crucial neoplastic and polyp categories

Jorge Gonzalez-Zapata, Francisco Lopez-Tiro, Elias Villalvazo-Avila et al.

Several Deep Learning (DL) methods have recently been proposed for an automated identification of kidney stones during an ureteroscopy to enable rapid therapeutic decisions. Even if these DL approaches led to promising results, they are mainly appropriate for kidney stone types for which numerous labelled data are available. However, only few labelled images are available for some rare kidney stone types. This contribution exploits Deep Metric Learning (DML) methods i) to handle such classes with few samples, ii) to generalize well to out of distribution samples, and iii) to cope better with new classes which are added to the database. The proposed Guided Deep Metric Learning approach is based on a novel architecture which was designed to learn data representations in an improved way. The solution was inspired by Few-Shot Learning (FSL) and makes use of a teacher-student approach. The teacher model (GEMINI) generates a reduced hypothesis space based on prior knowledge from the labeled data, and is used it as a guide to a student model (i.e., ResNet50) through a Knowledge Distillation scheme. Extensive tests were first performed on two datasets separately used for the recognition, namely a set of images acquired for the surfaces of the kidney stone fragments, and a set of images of the fragment sections. The proposed DML-approach improved the identification accuracy by 10% and 12% in comparison to DL-methods and other DML-approaches, respectively. Moreover, model embeddings from the two dataset types were merged in an organized way through a multi-view scheme to simultaneously exploit the information of surface and section fragments. Test with the resulting mixed model improves the identification accuracy by at least 3% and up to 30% with respect to DL-models and shallow machine learning methods, respectively.

Daniel Flores-Araiza, Francisco Lopez-Tiro, Elias Villalvazo-Avila et al.

Identifying the type of kidney stones can allow urologists to determine their formation cause, improving the early prescription of appropriate treatments to diminish future relapses. However, currently, the associated ex-vivo diagnosis (known as morpho-constitutional analysis, MCA) is time-consuming, expensive, and requires a great deal of experience, as it requires a visual analysis component that is highly operator dependant. Recently, machine learning methods have been developed for in-vivo endoscopic stone recognition. Shallow methods have been demonstrated to be reliable and interpretable but exhibit low accuracy, while deep learning-based methods yield high accuracy but are not explainable. However, high stake decisions require understandable computer-aided diagnosis (CAD) to suggest a course of action based on reasonable evidence, rather than merely prescribe one. Herein, we investigate means for learning part-prototypes (PPs) that enable interpretable models. Our proposal suggests a classification for a kidney stone patch image and provides explanations in a similar way as those used on the MCA method.

Francisco Lopez-Tiro, Elias Villalvazo-Avila, Juan Pablo Betancur-Rengifo et al.

This contribution presents a deep-learning method for extracting and fusing image information acquired from different viewpoints, with the aim to produce more discriminant object features for the identification of the type of kidney stones seen in endoscopic images. The model was further improved with a two-step transfer learning approach and by attention blocks to refine the learned feature maps. Deep feature fusion strategies improved the results of single view extraction backbone models by more than 6% in terms of accuracy of the kidney stones classification.

Jorge Gonzalez-Zapata, Ivan Reyes-Amezcua, Daniel Flores-Araiza et al.

Deep Metric Learning (DML) methods have been proven relevant for visual similarity learning. However, they sometimes lack generalization properties because they are trained often using an inappropriate sample selection strategy or due to the difficulty of the dataset caused by a distributional shift in the data. These represent a significant drawback when attempting to learn the underlying data manifold. Therefore, there is a pressing need to develop better ways of obtaining generalization and representation of the underlying manifold. In this paper, we propose a novel approach to DML that we call Guided Deep Metric Learning, a novel architecture oriented to learning more compact clusters, improving generalization under distributional shifts in DML. This novel architecture consists of two independent models: A multi-branch master model, inspired from a Few-Shot Learning (FSL) perspective, generates a reduced hypothesis space based on prior knowledge from labeled data, which guides or regularizes the decision boundary of a student model during training under an offline knowledge distillation scheme. Experiments have shown that the proposed method is capable of a better manifold generalization and representation to up to 40% improvement (Recall@1, CIFAR10), using guidelines suggested by Musgrave et al. to perform a more fair and realistic comparison, which is currently absent in the literature

Elias Villalvazo-Avila, Francisco Lopez-Tiro, Daniel Flores-Araiza et al.

This contribution presents a deep-learning method for extracting and fusing image information acquired from different viewpoints with the aim to produce more discriminant object features. Our approach was specifically designed to mimic the morpho-constitutional analysis used by urologists to visually classify kidney stones by inspecting the sections and surfaces of their fragments. Deep feature fusion strategies improved the results of single view extraction backbone models by more than 10\% in terms of precision of the kidney stones classification.

Ivan Reyes-Amezcua, Ricardo Espinosa, Christian Daul et al.

Accurate depth estimation in endoscopy is vital for successfully implementing computer vision pipelines for various medical procedures and CAD tools. In this paper, we present the EndoDepth benchmark, an evaluation framework designed to assess the robustness of monocular depth prediction models in endoscopic scenarios. Unlike traditional datasets, the EndoDepth benchmark incorporates common challenges encountered during endoscopic procedures. We present an evaluation approach that is consistent and specifically designed to evaluate the robustness performance of the model in endoscopic scenarios. Among these is a novel composite metric called the mean Depth Estimation Robustness Score (mDERS), which offers an in-depth evaluation of a model's accuracy against errors brought on by endoscopic image corruptions. Moreover, we present SCARED-C, a new dataset designed specifically to assess endoscopy robustness. Through extensive experimentation, we evaluate state-of-the-art depth prediction architectures on the EndoDepth benchmark, revealing their strengths and weaknesses in handling endoscopic challenging imaging artifacts. Our results demonstrate the importance of specialized techniques for accurate depth estimation in endoscopy and provide valuable insights for future research directions.

Carmina Pérez-Guerrero, Jorge Francisco Ciprián-Sánchez, Adriana Palacios et al.

Among the different kinds of fire accidents that can occur during industrial activities that involve hazardous materials, jet fires are one of the lesser-known types. This is because they are often involved in a process that generates a sequence of other accidents of greater magnitude, known as domino effect. Flame impingement usually causes domino effects, and jet fires present specific features that can significantly increase the probability of this happening. These features become relevant from a risk analysis perspective, making their proper characterization a crucial task. Deep Learning approaches have become extensively used for tasks such as jet fire characterization; however, these methods are heavily dependent on the amount of data and the quality of the labels. Data acquisition of jet fires involve expensive experiments, especially so if infrared imagery is used. Therefore, this paper proposes the use of Generative Adversarial Networks to produce plausible infrared images from visible ones, making experiments less expensive and allowing for other potential applications. The results suggest that it is possible to realistically replicate the results for experiments carried out using both visible and infrared cameras. The obtained results are compared with some previous experiments, and it is shown that similar results were obtained.

Ivan Reyes-Amezcua, Daniel Flores-Araiza, Gilberto Ochoa-Ruiz et al.

Feature engineering has become one of the most important steps to improve model prediction performance, and to produce quality datasets. However, this process requires non-trivial domain-knowledge which involves a time-consuming process. Thereby, automating such process has become an active area of research and of interest in industrial applications. In this paper, a novel method, called Meta-learning and Causality Based Feature Engineering (MACFE), is proposed; our method is based on the use of meta-learning, feature distribution encoding, and causality feature selection. In MACFE, meta-learning is used to find the best transformations, then the search is accelerated by pre-selecting "original" features given their causal relevance. Experimental evaluations on popular classification datasets show that MACFE can improve the prediction performance across eight classifiers, outperforms the current state-of-the-art methods in average by at least 6.54%, and obtains an improvement of 2.71% over the best previous works.

Gerardo Valente Vazquez-Garcia, Carmina Perez Guerrero, Eduardo Garduño et al.

In this work we present a novel framework for fire safety management in industrial settings through the implementation of a smart camera platform for jet flames characterization. The approach seeks to alleviate the lack of real-time solutions for industrial early fire segmentation and characterization. As a case study, we demonstrate how a SoC FPGA, running optimized Artificial Intelligence (AI) models can be leveraged to implement a full edge processing pipeline for jet flames analysis. In this paper we extend previous work on computer-vision jet fire segmentation by creating a novel experimental set-up and system implementation for addressing this issue, which can be replicated to other fire safety applications. The proposed platform is designed to carry out image processing tasks in real-time and on device, reducing video processing overheads, and thus the overall latency. This is achieved by optimizing a UNet segmentation model to make it amenable for an SoC FPGAs implementation; the optimized model can then be efficiently mapped onto the SoC reconfigurable logic for massively parallel execution. For our experiments, we have chosen the Ultra96 platform, as it also provides the means for implementing full-fledged intelligent systems using the SoC peripherals, as well as other Operating System (OS) capabilities (i.e., multi-threading) for systems management. For optimizing the model we made use of the Vitis (Xilinx) framework, which enabled us to optimize the full precision model from 7.5 million parameters to 59,095 parameters (125x less), which translated into a reduction of the processing latency of 2.9x. Further optimization (multi-threading and batch normalization) led to an improvement of 7.5x in terms of latency, yielding a performance of 30 Frames Per Second (FPS) without sacrificing accuracy in terms of the evaluated metrics (Dice Score).

Pablo Cesar Quihui-Rubio, Daniel Flores-Araiza, Gilberto Ochoa-Ruiz et al.

This study focuses on comparing deep learning methods for the segmentation and quantification of uncertainty in prostate segmentation from MRI images. The aim is to improve the workflow of prostate cancer detection and diagnosis. Seven different U-Net-based architectures, augmented with Monte-Carlo dropout, are evaluated for automatic segmentation of the central zone, peripheral zone, transition zone, and tumor, with uncertainty estimation. The top-performing model in this study is the Attention R2U-Net, achieving a mean Intersection over Union (IoU) of 76.3% and Dice Similarity Coefficient (DSC) of 85% for segmenting all zones. Additionally, Attention R2U-Net exhibits the lowest uncertainty values, particularly in the boundaries of the transition zone and tumor, when compared to the other models.

Pablo Cesar Quihui-Rubio, Gilberto Ochoa-Ruiz, Miguel Gonzalez-Mendoza et al.

Prostate cancer is the second-most frequently diagnosed cancer and the sixth leading cause of cancer death in males worldwide. The main problem that specialists face during the diagnosis of prostate cancer is the localization of Regions of Interest (ROI) containing a tumor tissue. Currently, the segmentation of this ROI in most cases is carried out manually by expert doctors, but the procedure is plagued with low detection rates (of about 27-44%) or overdiagnosis in some patients. Therefore, several research works have tackled the challenge of automatically segmenting and extracting features of the ROI from magnetic resonance images, as this process can greatly facilitate many diagnostic and therapeutic applications. However, the lack of clear prostate boundaries, the heterogeneity inherent to the prostate tissue, and the variety of prostate shapes makes this process very difficult to automate.In this work, six deep learning models were trained and analyzed with a dataset of MRI images obtained from the Centre Hospitalaire de Dijon and Universitat Politecnica de Catalunya. We carried out a comparison of multiple deep learning models (i.e. U-Net, Attention U-Net, Dense-UNet, Attention Dense-UNet, R2U-Net, and Attention R2U-Net) using categorical cross-entropy loss function. The analysis was performed using three metrics commonly used for image segmentation: Dice score, Jaccard index, and mean squared error. The model that give us the best result segmenting all the zones was R2U-Net, which achieved 0.869, 0.782, and 0.00013 for Dice, Jaccard and mean squared error, respectively.

Daniela Herrera, Gilberto Ochoa-Ruiz, Miguel Gonzalez-Mendoza et al.

The retinal vessel network studied through fundus images contributes to the diagnosis of multiple diseases not only found in the eye. The segmentation of this system may help the specialized task of analyzing these images by assisting in the quantification of morphological characteristics. Due to its relevance, several Deep Learning-based architectures have been tested for tackling this problem automatically. However, the impact of loss function selection on the segmentation of the intricate retinal blood vessel system hasn't been systematically evaluated. In this work, we present the comparison of the loss functions Binary Cross Entropy, Dice, Tversky, and Combo loss using the deep learning architectures (i.e. U-Net, Attention U-Net, and Nested UNet) with the DRIVE dataset. Their performance is assessed using four metrics: the AUC, the mean squared error, the dice score, and the Hausdorff distance. The models were trained with the same number of parameters and epochs. Using dice score and AUC, the best combination was SA-UNet with Combo loss, which had an average of 0.9442 and 0.809 respectively. The best average of Hausdorff distance and mean square error were obtained using the Nested U-Net with the Dice loss function, which had an average of 6.32 and 0.0241 respectively. The results showed that there is a significant difference in the selection of loss function

Ricardo Espinosa, Carlos Axel Garcia-Vega, Gilberto Ochoa-Ruiz et al.

This contribution shows how an appropriate image pre-processing can improve a deep-learning based 3D reconstruction of colon parts. The assumption is that, rather than global image illumination corrections, local under- and over-exposures should be corrected in colonoscopy. An overview of the pipeline including the image exposure correction and a RNN-SLAM is first given. Then, this paper quantifies the reconstruction accuracy of the endoscope trajectory in the colon with and without appropriate illumination correction

Pedro Esteban Chavarrias-Solano, Carlos Axel Garcia-Vega, Francisco Javier Lopez-Tiro et al.

In this contribution we use an ensemble deep-learning method for combining the prediction of two individual one-stage detectors (i.e., YOLOv4 and Yolact) with the aim to detect artefacts in endoscopic images. This ensemble strategy enabled us to improve the robustness of the individual models without harming their real-time computation capabilities. We demonstrated the effectiveness of our approach by training and testing the two individual models and various ensemble configurations on the "Endoscopic Artifact Detection Challenge" dataset. Extensive experiments show the superiority, in terms of mean average precision, of the ensemble approach over the individual models and previous works in the state of the art.

Ivan Reyes-Amezcua, Michael Rojas-Ruiz, Gilberto Ochoa-Ruiz et al.

Deep learning developments have improved medical imaging diagnoses dramatically, increasing accuracy in several domains. Nonetheless, obstacles continue to exist because of the requirement for huge datasets and legal limitations on data exchange. A solution is provided by Federated Learning (FL), which permits decentralized model training while maintaining data privacy. However, FL models are susceptible to data corruption, which may result in performance degradation. Using pre-trained models, this research suggests a strong FL framework to improve kidney stone diagnosis. Two different kidney stone datasets, each with six different categories of images, are used in our experimental setting. Our method involves two stages: Learning Parameter Optimization (LPO) and Federated Robustness Validation (FRV). We achieved a peak accuracy of 84.1% with seven epochs and 10 rounds during LPO stage, and 77.2% during FRV stage, showing enhanced diagnostic accuracy and robustness against image corruption. This highlights the potential of merging pre-trained models with FL to address privacy and performance concerns in medical diagnostics, and guarantees improved patient care and enhanced trust in FL-based medical systems.

Daniel Flores-Araiza, Francisco Lopez-Tiro, Clément Larose et al.

The in-vivo identification of the kidney stone types during an ureteroscopy would be a major medical advance in urology, as it could reduce the time of the tedious renal calculi extraction process, while diminishing infection risks. Furthermore, such an automated procedure would make possible to prescribe anti-recurrence treatments immediately. Nowadays, only few experienced urologists are able to recognize the kidney stone types in the images of the videos displayed on a screen during the endoscopy. Thus, several deep learning (DL) models have recently been proposed to automatically recognize the kidney stone types using ureteroscopic images. However, these DL models are of black box nature whicl limits their applicability in clinical settings. This contribution proposes a case-based reasoning DL model which uses prototypical parts (PPs) and generates local and global descriptors. The PPs encode for each class (i.e., kidney stone type) visual feature information (hue, saturation, intensity and textures) similar to that used by biologists. The PPs are optimally generated due a new loss function used during the model training. Moreover, the local and global descriptors of PPs allow to explain the decisions ("what" information, "where in the images") in an understandable way for biologists and urologists. The proposed DL model has been tested on a database including images of the six most widespread kidney stone types. The overall average classification accuracy was 90.37. When comparing this results with that of the eight other DL models of the kidney stone state-of-the-art, it can be seen that the valuable gain in explanability was not reached at the expense of accuracy which was even slightly increased with respect to that (88.2) of the best method of the literature. These promising and interpretable results also encourage urologists to put their trust in AI-based solutions.

Noha Ghatwary, Jiangbei Yue, Ahmed Elgendy et al.

Ulcerative colitis (UC) is a chronic mucosal inflammatory condition that places patients at increased risk of colorectal cancer. Colonoscopic surveillance remains the gold standard for assessing disease activity, and reporting typically relies on standardised endoscopic scoring metrics. The most widely used is the Mayo Endoscopic Score (MES), with some centres also adopting the Ulcerative Colitis Endoscopic Index of Severity (UCEIS). Both are descriptive assessments of mucosal inflammation (MES: 0 to 3; UCEIS: 0 to 8), where higher values indicate more severe disease. However, computational methods for automatically predicting these scores remain limited, largely due to the lack of publicly available expert-annotated datasets and the absence of robust benchmarking. There is also a significant research gap in generating clinically meaningful descriptions of UC images, despite image captioning being a well-established computer vision task. Variability in endoscopic systems and procedural workflows across centres further highlights the need for multi-centre datasets to ensure algorithmic robustness and generalisability. In this work, we introduce a curated multi-centre, multi-resolution dataset that includes expert-validated MES and UCEIS labels, alongside detailed clinical descriptions. To our knowledge, this is the first comprehensive dataset that combines dual scoring metrics for classification tasks with expert-generated captions describing mucosal appearance and clinically accepted reasoning for image captioning. This resource opens new opportunities for developing clinically meaningful multimodal algorithms. In addition to the dataset, we also provide benchmarking using convolutional neural networks, vision transformers, hybrid models, and widely used multimodal vision-language captioning algorithms.

Ivan Reyes-Amezcua, Francisco Lopez-Tiro, Clément Larose et al.

The reliability of artificial intelligence (AI) in medical imaging critically depends on its robustness to heterogeneous and corrupted images acquired with diverse devices across different hospitals which is highly challenging. Therefore, this paper introduces FedAgain, a trust-based Federated Learning (Federated Learning) strategy designed to enhance robustness and generalization for automated kidney stone identification from endoscopic images. FedAgain integrates a dual trust mechanism that combines benchmark reliability and model divergence to dynamically weight client contributions, mitigating the impact of noisy or adversarial updates during aggregation. The framework enables the training of collaborative models across multiple institutions while preserving data privacy and promoting stable convergence under real-world conditions. Extensive experiments across five datasets, including two canonical benchmarks (MNIST and CIFAR-10), two private multi-institutional kidney stone datasets, and one public dataset (MyStone), demonstrate that FedAgain consistently outperforms standard Federated Learning baselines under non-identically and independently distributed (non-IID) data and corrupted-client scenarios. By maintaining diagnostic accuracy and performance stability under varying conditions, FedAgain represents a practical advance toward reliable, privacy-preserving, and clinically deployable federated AI for medical imaging.

Ruben Gonzalez-Perez, Francisco Lopez-Tiro, Ivan Reyes-Amezcua et al.

Currently, the Morpho-Constitutional Analysis (MCA) is the de facto approach for the etiological diagnosis of kidney stone formation, and it is an important step for establishing personalized treatment to avoid relapses. More recently, research has focused on performing such tasks intra-operatively, an approach known as Endoscopic Stone Recognition (ESR). Both methods rely on features observed in the surface and the section of kidney stones to separate the analyzed samples into several sub-groups. However, given the high intra-observer variability and the complex operating conditions found in ESR, there is a lot of interest in using AI for computer-aided diagnosis. However, current AI models require large datasets to attain a good performance and for generalizing to unseen distributions. This is a major problem as large labeled datasets are very difficult to acquire, and some classes of kidney stones are very rare. Thus, in this paper, we present a method based on diffusion as a way of augmenting pre-existing ex-vivo kidney stone datasets. Our aim is to create plausible diverse kidney stone images that can be used for pre-training models using ex-vivo data. We show that by mixing natural and synthetic images of CCD images, it is possible to train models capable of performing very well on unseen intra-operative data. Our results show that is possible to attain an improvement of 10% in terms of accuracy compared to a baseline model pre-trained only on ImageNet. Moreover, our results show an improvement of 6% for surface images and 10% for section images compared to a model train on CCD images only, which demonstrates the effectiveness of using synthetic images.

Mansoor Ali Teevno, Rafael Martinez-Garcia-Pena, Gilberto Ochoa-Ruiz et al.

Frequent monitoring is necessary to stratify individuals based on their likelihood of developing gastrointestinal (GI) cancer precursors. In clinical practice, white-light imaging (WLI) and complementary modalities such as narrow-band imaging (NBI) and fluorescence imaging are used to assess risk areas. However, conventional deep learning (DL) models show degraded performance due to the domain gap when a model is trained on one modality and tested on a different one. In our earlier approach, we used a superpixel-based method referred to as "SUPRA" to effectively learn domain-invariant information using color and space distances to generate groups of pixels. One of the main limitations of this earlier work is that the aggregation does not exploit structural information, making it suboptimal for segmentation tasks, especially for polyps and heterogeneous color distributions. Therefore, in this work, we propose an approach for style-content disentanglement using instance normalization and instance selective whitening (ISW) for improved domain generalization when combined with SUPRA. We evaluate our approach on two datasets: EndoUDA Barrett's Esophagus and EndoUDA polyps, and compare its performance with three state-of-the-art (SOTA) methods. Our findings demonstrate a notable enhancement in performance compared to both baseline and SOTA methods across the target domain data. Specifically, our approach exhibited improvements of 14%, 10%, 8%, and 18% over the baseline and three SOTA methods on the polyp dataset. Additionally, it surpassed the second-best method (EndoUDA) on the Barrett's Esophagus dataset by nearly 2%.

Mansoor Ali, Maksim Richards, Gilberto Ochoa-Ruiz et al.

While recent advances in deep learning for surgical scene segmentation have demonstrated promising results on single-centre and single-imaging modality data, these methods usually do not generalise to unseen distribution (i.e., from other centres) and unseen modalities. Current literature for tackling generalisation on out-of-distribution data and domain gaps due to modality changes has been widely researched but mostly for natural scene data. However, these methods cannot be directly applied to the surgical scenes due to limited visual cues and often extremely diverse scenarios compared to the natural scene data. Inspired by these works in natural scenes to push generalisability on OOD data, we hypothesise that exploiting the style and content information in the surgical scenes could minimise the appearances, making it less variable to sudden changes such as blood or imaging artefacts. This can be achieved by performing instance normalisation and feature covariance mapping techniques for robust and generalisable feature representations. Further, to eliminate the risk of removing salient feature representation associated with the objects of interest, we introduce a restitution module within the feature learning ResNet backbone that can enable the retention of useful task-relevant features. To tackle the lack of multiclass and multicentre data for surgical scene segmentation, we also provide a newly curated dataset that can be vital for addressing generalisability in this domain. Our proposed RobustSurg obtained nearly 23% improvement on the baseline DeepLabv3+ and from 10-32% improvement on the SOTA in terms of mean IoU score on an unseen centre HeiCholSeg dataset when trained on CholecSeg8K. Similarly, RobustSurg also obtained nearly 22% improvement over the baseline and nearly 11% improvement on a recent SOTA method for the target set of the EndoUDA polyp dataset.

Ivan Reyes-Amezcua, Francisco Lopez-Tiro, Clement Larose et al.

Kidney stone classification from endoscopic images is critical for personalized treatment and recurrence prevention. While convolutional neural networks (CNNs) have shown promise in this task, their limited ability to capture long-range dependencies can hinder performance under variable imaging conditions. This study presents a comparative analysis between Vision Transformers (ViTs) and CNN-based models, evaluating their performance on two ex vivo datasets comprising CCD camera and flexible ureteroscope images. The ViT-base model pretrained on ImageNet-21k consistently outperformed a ResNet50 baseline across multiple imaging conditions. For instance, in the most visually complex subset (Section patches from endoscopic images), the ViT model achieved 95.2% accuracy and 95.1% F1-score, compared to 64.5% and 59.3% with ResNet50. In the mixed-view subset from CCD-camera images, ViT reached 87.1% accuracy versus 78.4% with CNN. These improvements extend across precision and recall as well. The results demonstrate that ViT-based architectures provide superior classification performance and offer a scalable alternative to conventional CNNs for kidney stone image analysis.

Carlos Salazar-Ruiz, Francisco Lopez-Tiro, Ivan Reyes-Amezcua et al.

Determining the type of kidney stones is crucial for prescribing appropriate treatments to prevent recurrence. Currently, various approaches exist to identify the type of kidney stones. However, obtaining results through the reference ex vivo identification procedure can take several weeks, while in vivo visual recognition requires highly trained specialists. For this reason, deep learning models have been developed to provide urologists with an automated classification of kidney stones during ureteroscopies. Nevertheless, a common issue with these models is the lack of training data. This contribution presents a deep learning method based on few-shot learning, aimed at producing sufficiently discriminative features for identifying kidney stone types in endoscopic images, even with a very limited number of samples. This approach was specifically designed for scenarios where endoscopic images are scarce or where uncommon classes are present, enabling classification even with a limited training dataset. The results demonstrate that Prototypical Networks, using up to 25% of the training data, can achieve performance equal to or better than traditional deep learning models trained with the complete dataset.

Martin Villagrana, Francisco Lopez-Tiro, Clement Larose et al.

The segmentation of kidney stones is regarded as a critical preliminary step to enable the identification of urinary stone types through machine- or deep-learning-based approaches. In urology, manual segmentation is considered tedious and impractical due to the typically large scale of image databases and the continuous generation of new data. In this study, the potential of the Segment Anything Model (SAM) -- a state-of-the-art deep learning framework -- is investigated for the automation of kidney stone segmentation. The performance of SAM is evaluated in comparison to traditional models, including U-Net, Residual U-Net, and Attention U-Net, which, despite their efficiency, frequently exhibit limitations in generalizing to unseen datasets. The findings highlight SAM's superior adaptability and efficiency. While SAM achieves comparable performance to U-Net on in-distribution data (Accuracy: 97.68 + 3.04; Dice: 97.78 + 2.47; IoU: 95.76 + 4.18), it demonstrates significantly enhanced generalization capabilities on out-of-distribution data, surpassing all U-Net variants by margins of up to 23 percent.

Mansoor Ali Teevno, Gilberto Ochoa-Ruiz, Sharib Ali

While recent advances in deep learning (DL) for surgical scene segmentation have yielded promising results on single-center and single-imaging modality data, these methods usually do not generalize well to unseen distributions or modalities. Even though human experts can identify visual appearances, DL methods often fail to do so when data samples do not follow a similar distribution. Current literature addressing domain gaps in modality changes has focused primarily on natural scene data. However, these methods cannot be directly applied to endoscopic data, as visual cues in such data are more limited compared to natural scenes. In this work, we exploit both style and content information in images by performing instance normalization and feature covariance mapping techniques to preserve robust and generalizable feature representations. Additionally, to avoid the risk of removing salient feature representations associated with objects of interest, we introduce a restitution module within the feature-learning ResNet backbone that retains useful task-relevant features. Our proposed method shows a 13.7% improvement over the baseline DeepLabv3+ and nearly an 8% improvement over recent state-of-the-art (SOTA) methods for the target (different modality) set of the EndoUDA polyp dataset. Similarly, our method achieved a 19% improvement over the baseline and 6% over the best-performing SOTA method on the EndoUDA Barrett's esophagus (BE) dataset.

Pablo Cesar Quihui-Rubio, Daniel Flores-Araiza, Miguel Gonzalez-Mendoza et al.

This contribution presents a deep learning method for the segmentation of prostate zones in MRI images based on U-Net using additive and feature pyramid attention modules, which can improve the workflow of prostate cancer detection and diagnosis. The proposed model is compared to seven different U-Net-based architectures. The automatic segmentation performance of each model of the central zone (CZ), peripheral zone (PZ), transition zone (TZ) and Tumor were evaluated using Dice Score (DSC), and the Intersection over Union (IoU) metrics. The proposed alternative achieved a mean DSC of 84.15% and IoU of 76.9% in the test set, outperforming most of the studied models in this work except from R2U-Net and attention R2U-Net architectures.

Eduardo Guarduño-Martinez, Jorge Ciprian-Sanchez, Gerardo Valente et al.

Wildfires represent one of the most relevant natural disasters worldwide, due to their impact on various societal and environmental levels. Thus, a significant amount of research has been carried out to investigate and apply computer vision techniques to address this problem. One of the most promising approaches for wildfire fighting is the use of drones equipped with visible and infrared cameras for the detection, monitoring, and fire spread assessment in a remote manner but in close proximity to the affected areas. However, implementing effective computer vision algorithms on board is often prohibitive since deploying full-precision deep learning models running on GPU is not a viable option, due to their high power consumption and the limited payload a drone can handle. Thus, in this work, we posit that smart cameras, based on low-power consumption field-programmable gate arrays (FPGAs), in tandem with binarized neural networks (BNNs), represent a cost-effective alternative for implementing onboard computing on the edge. Herein we present the implementation of a segmentation model applied to the Corsican Fire Database. We optimized an existing U-Net model for such a task and ported the model to an edge device (a Xilinx Ultra96-v2 FPGA). By pruning and quantizing the original model, we reduce the number of parameters by 90%. Furthermore, additional optimizations enabled us to increase the throughput of the original model from 8 frames per second (FPS) to 33.63 FPS without loss in the segmentation performance: our model obtained 0.912 in Matthews correlation coefficient (MCC),0.915 in F1 score and 0.870 in Hafiane quality index (HAF), and comparable qualitative segmentation results when contrasted to the original full-precision model. The final model was integrated into a low-cost FPGA, which was used to implement a neural network accelerator.

Mauricio Mendez-Ruiz, Jorge Gonzalez-Zapata, Ivan Reyes-Amezcua et al.

Few-shot learning is a challenging area of research that aims to learn new concepts with only a few labeled samples of data. Recent works based on metric-learning approaches leverage the meta-learning approach, which is encompassed by episodic tasks that make use a support (training) and query set (test) with the objective of learning a similarity comparison metric between those sets. Due to the lack of data, the learning process of the embedding network becomes an important part of the few-shot task. Previous works have addressed this problem using metric learning approaches, but the properties of the underlying latent space and the separability of the difference classes on it was not entirely enforced. In this work, we propose two different loss functions which consider the importance of the embedding vectors by looking at the intra-class and inter-class distance between the few data. The first loss function is the Proto-Triplet Loss, which is based on the original triplet loss with the modifications needed to better work on few-shot scenarios. The second loss function, which we dub ICNN loss is based on an inter and intra class nearest neighbors score, which help us to assess the quality of embeddings obtained from the trained network. Our results, obtained from a extensive experimental setup show a significant improvement in accuracy in the miniImagenNet benchmark compared to other metric-based few-shot learning methods by a margin of 2%, demonstrating the capability of these loss functions to allow the network to generalize better to previously unseen classes. In our experiments, we demonstrate competitive generalization capabilities to other domains, such as the Caltech CUB, Dogs and Cars datasets compared with the state of the art.

Francisco Lopez-Tiro, Vincent Estrade, Jacques Hubert et al.

Determining the type of kidney stones allows urologists to prescribe a treatment to avoid recurrence of renal lithiasis. An automated in-vivo image-based classification method would be an important step towards an immediate identification of the kidney stone type required as a first phase of the diagnosis. In the literature it was shown on ex-vivo data (i.e., in very controlled scene and image acquisition conditions) that an automated kidney stone classification is indeed feasible. This pilot study compares the kidney stone recognition performances of six shallow machine learning methods and three deep-learning architectures which were tested with in-vivo images of the four most frequent urinary calculi types acquired with an endoscope during standard ureteroscopies. This contribution details the database construction and the design of the tested kidney stones classifiers. Even if the best results were obtained by the Inception v3 architecture (weighted precision, recall and F1-score of 0.97, 0.98 and 0.97, respectively), it is also shown that choosing an appropriate colour space and texture features allows a shallow machine learning method to approach closely the performances of the most promising deep-learning methods (the XGBoost classifier led to weighted precision, recall and F1-score values of 0.96). This paper is the first one that explores the most discriminant features to be extracted from images acquired during ureteroscopies.

Carmina Pérez-Guerrero, Adriana Palacios, Gilberto Ochoa-Ruiz et al.

Jet fires are relatively small and have the least severe effects among the diverse fire accidents that can occur in industrial plants; however, they are usually involved in a process known as the domino effect, that leads to more severe events, such as explosions or the initiation of another fire, making the analysis of such fires an important part of risk analysis. This research work explores the application of deep learning models in an alternative approach that uses the semantic segmentation of jet fires flames to extract main geometrical attributes, relevant for fire risk assessments. A comparison is made between traditional image processing methods and some state-of-the-art deep learning models. It is found that the best approach is a deep learning architecture known as UNet, along with its two improvements, Attention UNet and UNet++. The models are then used to segment a group of vertical jet flames of varying pipe outlet diameters to extract their main geometrical characteristics. Attention UNet obtained the best general performance in the approximation of both height and area of the flames, while also showing a statistically significant difference between it and UNet++. UNet obtained the best overall performance for the approximation of the lift-off distances; however, there is not enough data to prove a statistically significant difference between Attention UNet and UNet++. The only instance where UNet++ outperformed the other models, was while obtaining the lift-off distances of the jet flames with 0.01275 m pipe outlet diameter. In general, the explored models show good agreement between the experimental and predicted values for relatively large turbulent propane jet flames, released in sonic and subsonic regimes; thus, making these radiation zones segmentation models, a suitable approach for different jet flame risk management scenarios.

Juan Carlos Angeles-Ceron, Gilberto Ochoa-Ruiz, Leonardo Chang et al.

Precise instrument segmentation aid surgeons to navigate the body more easily and increase patient safety. While accurate tracking of surgical instruments in real-time plays a crucial role in minimally invasive computer-assisted surgeries, it is a challenging task to achieve, mainly due to 1) complex surgical environment, and 2) model design with both optimal accuracy and speed. Deep learning gives us the opportunity to learn complex environment from large surgery scene environments and placements of these instruments in real world scenarios. The Robust Medical Instrument Segmentation 2019 challenge (ROBUST-MIS) provides more than 10,000 frames with surgical tools in different clinical settings. In this paper, we use a light-weight single stage instance segmentation model complemented with a convolutional block attention module for achieving both faster and accurate inference. We further improve accuracy through data augmentation and optimal anchor localisation strategies. To our knowledge, this is the first work that explicitly focuses on both real-time performance and improved accuracy. Our approach out-performed top team performances in the ROBUST-MIS challenge with over 44% improvement on both area-based metric MI_DSC and distance-based metric MI_NSD. We also demonstrate real-time performance (> 60 frames-per-second) with different but competitive variants of our final approach.

Carmina Pérez-Guerrero, Adriana Palacios, Gilberto Ochoa-Ruiz et al.

Risk assessment is relevant in any workplace, however there is a degree of unpredictability when dealing with flammable or hazardous materials so that detection of fire accidents by itself may not be enough. An example of this is the impingement of jet fires, where the heat fluxes of the flame could reach nearby equipment and dramatically increase the probability of a domino effect with catastrophic results. Because of this, the characterization of such fire accidents is important from a risk management point of view. One such characterization would be the segmentation of different radiation zones within the flame, so this paper presents an exploratory research regarding several traditional computer vision and Deep Learning segmentation approaches to solve this specific problem. A data set of propane jet fires is used to train and evaluate the different approaches and given the difference in the distribution of the zones and background of the images, different loss functions, that seek to alleviate data imbalance, are also explored. Additionally, different metrics are correlated to a manual ranking performed by experts to make an evaluation that closely resembles the expert's criteria. The Hausdorff Distance and Adjusted Random Index were the metrics with the highest correlation and the best results were obtained from the UNet architecture with a Weighted Cross-Entropy Loss. These results can be used in future research to extract more geometric information from the segmentation masks or could even be implemented on other types of fire accidents.

Mauricio Mendez-Ruiz, Ivan Garcia Jorge Gonzalez-Zapata, Gilberto Ochoa-Ruiz et al.

Few-shot learning is a relatively new technique that specializes in problems where we have little amounts of data. The goal of these methods is to classify categories that have not been seen before with just a handful of samples. Recent approaches, such as metric learning, adopt the meta-learning strategy in which we have episodic tasks conformed by support (training) data and query (test) data. Metric learning methods have demonstrated that simple models can achieve good performance by learning a similarity function to compare the support and the query data. However, the feature space learned by a given metric learning approach may not exploit the information given by a specific few-shot task. In this work, we explore the use of dimension reduction techniques as a way to find task-significant features helping to make better predictions. We measure the performance of the reduced features by assigning a score based on the intra-class and inter-class distance, and selecting a feature reduction method in which instances of different classes are far away and instances of the same class are close. This module helps to improve the accuracy performance by allowing the similarity function, given by the metric learning method, to have more discriminative features for the classification. Our method outperforms the metric learning baselines in the miniImageNet dataset by around 2% in accuracy performance.

Cuauhtemoc Daniel Suarez-Ramirez, Miguel Gonzalez-Mendoza, Leonardo Chang-Fernandez et al.

The optimization of Binary Neural Networks (BNNs) relies on approximating the real-valued weights with their binarized representations. Current techniques for weight-updating use the same approaches as traditional Neural Networks (NNs) with the extra requirement of using an approximation to the derivative of the sign function - as it is the Dirac-Delta function - for back-propagation; thus, efforts are focused adapting full-precision techniques to work on BNNs. In the literature, only one previous effort has tackled the problem of directly training the BNNs with bit-flips by using the first raw moment estimate of the gradients and comparing it against a threshold for deciding when to flip a weight (Bop). In this paper, we take an approach parallel to Adam which also uses the second raw moment estimate to normalize the first raw moment before doing the comparison with the threshold, we call this method Bop2ndOrder. We present two versions of the proposed optimizer: a biased one and a bias-corrected one, each with its own applications. Also, we present a complete ablation study of the hyperparameters space, as well as the effect of using schedulers on each of them. For these studies, we tested the optimizer in CIFAR10 using the BinaryNet architecture. Also, we tested it in ImageNet 2012 with the XnorNet and BiRealNet architectures for accuracy. In both datasets our approach proved to converge faster, was robust to changes of the hyperparameters, and achieved better accuracy values.

Juan Carlos Angeles Ceron, Leonardo Chang, Gilberto Ochoa-Ruiz et al.

Image-based tracking of laparoscopic instruments plays a fundamental role in computer and robotic-assisted surgeries by aiding surgeons and increasing patient safety. Computer vision contests, such as the Robust Medical Instrument Segmentation (ROBUST-MIS) Challenge, seek to encourage the development of robust models for such purposes, providing large, diverse, and annotated datasets. To date, most of the existing models for instance segmentation of medical instruments were based on two-stage detectors, which provide robust results but are nowhere near to the real-time (5 frames-per-second (fps)at most). However, in order for the method to be clinically applicable, real-time capability is utmost required along with high accuracy. In this paper, we propose the addition of attention mechanisms to the YOLACT architecture that allows real-time instance segmentation of instrument with improved accuracy on the ROBUST-MIS dataset. Our proposed approach achieves competitive performance compared to the winner ofthe 2019 ROBUST-MIS challenge in terms of robustness scores,obtaining 0.313 MI_DSC and 0.338 MI_NSD, while achieving real-time performance (37 fps)