Michel de Mathelin

Jorge F. Lazo, Chun-Feng Lai, Sara Moccia et al.

Navigation inside luminal organs is an arduous task that requires non-intuitive coordination between the movement of the operator's hand and the information obtained from the endoscopic video. The development of tools to automate certain tasks could alleviate the physical and mental load of doctors during interventions, allowing them to focus on diagnosis and decision-making tasks. In this paper, we present a synergic solution for intraluminal navigation consisting of a 3D printed endoscopic soft robot that can move safely inside luminal structures. Visual servoing, based on Convolutional Neural Networks (CNNs) is used to achieve the autonomous navigation task. The CNN is trained with phantoms and in-vivo data to segment the lumen, and a model-less approach is presented to control the movement in constrained environments. The proposed robot is validated in anatomical phantoms in different path configurations. We analyze the movement of the robot using different metrics such as task completion time, smoothness, error in the steady-state, and mean and maximum error. We show that our method is suitable to navigate safely in hollow environments and conditions which are different than the ones the network was originally trained on.

Jorge F. Lazo, Benoit Rosa, Michele Catellani et al.

Objective: Accurate visual classification of bladder tissue during Trans-Urethral Resection of Bladder Tumor (TURBT) procedures is essential to improve early cancer diagnosis and treatment. During TURBT interventions, White Light Imaging (WLI) and Narrow Band Imaging (NBI) techniques are used for lesion detection. Each imaging technique provides diverse visual information that allows clinicians to identify and classify cancerous lesions. Computer vision methods that use both imaging techniques could improve endoscopic diagnosis. We address the challenge of tissue classification when annotations are available only in one domain, in our case WLI, and the endoscopic images correspond to an unpaired dataset, i.e. there is no exact equivalent for every image in both NBI and WLI domains. Method: We propose a semi-surprised Generative Adversarial Network (GAN)-based method composed of three main components: a teacher network trained on the labeled WLI data; a cycle-consistency GAN to perform unpaired image-to-image translation, and a multi-input student network. To ensure the quality of the synthetic images generated by the proposed GAN we perform a detailed quantitative, and qualitative analysis with the help of specialists. Conclusion: The overall average classification accuracy, precision, and recall obtained with the proposed method for tissue classification are 0.90, 0.88, and 0.89 respectively, while the same metrics obtained in the unlabeled domain (NBI) are 0.92, 0.64, and 0.94 respectively. The quality of the generated images is reliable enough to deceive specialists. Significance: This study shows the potential of using semi-supervised GAN-based bladder tissue classification when annotations are limited in multi-domain data. The dataset is available at https://zenodo.org/record/7741476#.ZBQUK7TMJ6k

Rongrong Liu, John M. Wandeto, Florent Nageotte et al.

This paper builds on our previous work by exploiting Artificial Intelligence to predict individual grip force variability in manual robot control. Grip forces were recorded from various loci in the dominant and non dominant hands of individuals by means of wearable wireless sensor technology. Statistical analyses bring to the fore skill specific temporal variations in thousands of grip forces of a complete novice and a highly proficient expert in manual robot control. A brain inspired neural network model that uses the output metric of a Self Organizing Map with unsupervised winner take all learning was run on the sensor output from both hands of each user. The neural network metric expresses the difference between an input representation and its model representation at any given moment in time t and reliably captures the differences between novice and expert performance in terms of grip force variability.Functionally motivated spatiotemporal analysis of individual average grip forces, computed for time windows of constant size in the output of a restricted amount of task-relevant sensors in the dominant (preferred) hand, reveal finger-specific synergies reflecting robotic task skill. The analyses lead the way towards grip force monitoring in real time to permit tracking task skill evolution in trainees, or identify individual proficiency levels in human robot interaction in environmental contexts of high sensory uncertainty. Parsimonious Artificial Intelligence (AI) assistance will contribute to the outcome of new types of surgery, in particular single-port approaches such as NOTES (Natural Orifice Transluminal Endoscopic Surgery) and SILS (Single Incision Laparoscopic Surgery).

Jorge F. Lazo, Sara Moccia, Aldo Marzullo et al.

Ureteroscopy and cystoscopy are the gold standard methods to identify and treat tumors along the urinary tract. It has been reported that during a normal procedure a rate of 10-20 % of the lesions could be missed. In this work we study the implementation of 3 different Convolutional Neural Networks (CNNs), using a 2-steps training strategy, to classify images from the urinary tract with and without lesions. A total of 6,101 images from ureteroscopy and cystoscopy procedures were collected. The CNNs were trained and tested using transfer learning in a two-steps fashion on 3 datasets. The datasets used were: 1) only ureteroscopy images, 2) only cystoscopy images and 3) the combination of both of them. For cystoscopy data, VGG performed better obtaining an Area Under the ROC Curve (AUC) value of 0.846. In the cases of ureteroscopy and the combination of both datasets, ResNet50 achieved the best results with AUC values of 0.987 and 0.940. The use of a training dataset that comprehends both domains results in general better performances, but performing a second stage of transfer learning achieves comparable ones. There is no single model which performs better in all scenarios, but ResNet50 is the network that achieves the best performances in most of them. The obtained results open the opportunity for further investigation with a view for improving lesion detection in endoscopic images of the urinary system.

Jorge F. Lazo, Aldo Marzullo, Sara Moccia et al.

Purpose: Ureteroscopy is an efficient endoscopic minimally invasive technique for the diagnosis and treatment of upper tract urothelial carcinoma (UTUC). During ureteroscopy, the automatic segmentation of the hollow lumen is of primary importance, since it indicates the path that the endoscope should follow. In order to obtain an accurate segmentation of the hollow lumen, this paper presents an automatic method based on Convolutional Neural Networks (CNNs). Methods: The proposed method is based on an ensemble of 4 parallel CNNs to simultaneously process single and multi-frame information. Of these, two architectures are taken as core-models, namely U-Net based in residual blocks($m_1$) and Mask-RCNN($m_2$), which are fed with single still-frames $I(t)$. The other two models ($M_1$, $M_2$) are modifications of the former ones consisting on the addition of a stage which makes use of 3D Convolutions to process temporal information. $M_1$, $M_2$ are fed with triplets of frames ($I(t-1)$, $I(t)$, $I(t+1)$) to produce the segmentation for $I(t)$. Results: The proposed method was evaluated using a custom dataset of 11 videos (2,673 frames) which were collected and manually annotated from 6 patients. We obtain a Dice similarity coefficient of 0.80, outperforming previous state-of-the-art methods. Conclusion: The obtained results show that spatial-temporal information can be effectively exploited by the ensemble model to improve hollow lumen segmentation in ureteroscopic images. The method is effective also in presence of poor visibility, occasional bleeding, or specular reflections.

Rongrong Liu, Florent Nageotte, Philippe Zanne et al.

Deep learning has provided new ways of manipulating, processing and analyzing data. It sometimes may achieve results comparable to, or surpassing human expert performance, and has become a source of inspiration in the era of artificial intelligence. Another subfield of machine learning named reinforcement learning, tries to find an optimal behavior strategy through interactions with the environment. Combining deep learning and reinforcement learning permits resolving critical issues relative to the dimensionality and scalability of data in tasks with sparse reward signals, such as robotic manipulation and control tasks, that neither method permits resolving when applied on its own. In this paper, we present recent significant progress of deep reinforcement learning algorithms, which try to tackle the problems for the application in the domain of robotic manipulation control, such as sample efficiency and generalization. Despite these continuous improvements, currently, the challenges of learning robust and versatile manipulation skills for robots with deep reinforcement learning are still far from being resolved for real world applications.

Rongrong Liu, Florent Nageotte, Philippe Zanne et al.

Wearable biosensor technology enables real-time, convenient, and continuous monitoring of users behavioral signals. Such include signals relative to body motion, body temperature, biological or biochemical markers, and individual grip forces, which are studied in this paper. A four step pick and drop image guided and robot assisted precision task has been designed for exploiting a wearable wireless sensor glove system. Individual spatio temporal grip forces are analyzed on the basis of thousands of individual sensor data, collected from different locations on the dominant and non-dominant hands of each of three users in ten successive task sessions. Statistical comparisons reveal specific differences between grip force profiles of the individual users as a function of task skill level (expertise) and time.

Jorge F. Lazo, Aldo Marzullo, Sara Moccia et al.

Ureteroscopy is becoming the first surgical treatment option for the majority of urinary affections. This procedure is performed using an endoscope which provides the surgeon with the visual information necessary to navigate inside the urinary tract. Having in mind the development of surgical assistance systems, that could enhance the performance of surgeon, the task of lumen segmentation is a fundamental part since this is the visual reference which marks the path that the endoscope should follow. This is something that has not been analyzed in ureteroscopy data before. However, this task presents several challenges given the image quality and the conditions itself of ureteroscopy procedures. In this paper, we study the implementation of a Deep Neural Network which exploits the advantage of residual units in an architecture based on U-Net. For the training of these networks, we analyze the use of two different color spaces: gray-scale and RGB data images. We found that training on gray-scale images gives the best results obtaining mean values of Dice Score, Precision, and Recall of 0.73, 0.58, and 0.92 respectively. The results obtained shows that the use of residual U-Net could be a suitable model for further development for a computer-aided system for navigation and guidance through the urinary system.

Birgitta Dresp-Langley, Florent Nageotte, Philippe Zanne et al.

Wearable sensor systems with transmitting capabilities are currently employed for the biometric screening of exercise activities and other performance data. Such technology is generally wireless and enables the noninvasive monitoring of signals to track and trace user behaviors in real time. Examples include signals relative to hand and finger movements or force control reflected by individual grip force data. As will be shown here, these signals directly translate into task, skill, and hand specific, dominant versus non dominant hand, grip force profiles for different measurement loci in the fingers and palm of the hand. The present study draws from thousands of such sensor data recorded from multiple spatial locations. The individual grip force profiles of a highly proficient left handed exper, a right handed dominant hand trained user, and a right handed novice performing an image guided, robot assisted precision task with the dominant or the non dominant hand are analyzed. The step by step statistical approach follows Tukeys detective work principle, guided by explicit functional assumptions relating to somatosensory receptive field organization in the human brain. Correlation analyses in terms of Person Product Moments reveal skill specific differences in covariation patterns in the individual grip force profiles. These can be functionally mapped to from global to local coding principles in the brain networks that govern grip force control and its optimization with a specific task expertise. Implications for the real time monitoring of grip forces and performance training in complex task user systems are brought forward.

Michel de Mathelin, Florent Nageotte, Philippe Zanne et al.

STRAS (Single access Transluminal Robotic Assistant for Surgeons) is a new robotic system for application to intraluminal surgical procedures. Preclinical testing of STRAS has recently permitted to demonstrate major advantages of the system in comparison with classic procedures. Benchmark methods permitting to establish objective criteria for expertise need to be worked out now to effectively train surgeons on this new system in the near future. STRAS consists of three cable driven subsystems, one endoscope serving as guide, and two flexible instruments. The flexible instruments have three degrees of freedom and can be teleoperated by a single user via two specially designed master interfaces. In this study here, small force sensors sewn into a wearable glove to ergonomically fit the master handles of the robotic system were employed for monitoring the forces applied by an expert and a trainee who was a complete novice during all the steps of surgical task execution in a simulator task, a four step pick and drop. Analysis of gripforce profiles is performed sensor by sensor to bring to the fore specific differences in handgrip force profiles in specific sensor locations on anatomically relevant parts of the fingers and hand controlling the master slave system.

Vinkle Srivastav, Thibaut Issenhuth, Abdolrahim Kadkhodamohammadi et al.

Person detection and pose estimation is a key requirement to develop intelligent context-aware assistance systems. To foster the development of human pose estimation methods and their applications in the Operating Room (OR), we release the Multi-View Operating Room (MVOR) dataset, the first public dataset recorded during real clinical interventions. It consists of 732 synchronized multi-view frames recorded by three RGB-D cameras in a hybrid OR. It also includes the visual challenges present in such environments, such as occlusions and clutter. We provide camera calibration parameters, color and depth frames, human bounding boxes, and 2D/3D pose annotations. In this paper, we present the dataset, its annotations, as well as baseline results from several recent person detection and 2D/3D pose estimation methods. Since we need to blur some parts of the images to hide identity and nudity in the released dataset, we also present a comparative study of how the baselines have been impacted by the blurring. Results show a large margin for improvement and suggest that the MVOR dataset can be useful to compare the performance of the different methods.

Abdolrahim Kadkhodamohammadi, Afshin Gangi, Michel de Mathelin et al.

Many approaches have been proposed for human pose estimation in single and multi-view RGB images. However, some environments, such as the operating room, are still very challenging for state-of-the-art RGB methods. In this paper, we propose an approach for multi-view 3D human pose estimation from RGB-D images and demonstrate the benefits of using the additional depth channel for pose refinement beyond its use for the generation of improved features. The proposed method permits the joint detection and estimation of the poses without knowing a priori the number of persons present in the scene. We evaluate this approach on a novel multi-view RGB-D dataset acquired during live surgeries and annotated with ground truth 3D poses.

Andru P. Twinanda, Didier Mutter, Jacques Marescaux et al.

The tool presence detection challenge at M2CAI 2016 consists of identifying the presence/absence of seven surgical tools in the images of cholecystectomy videos. Here, we propose to use deep architectures that are based on our previous work where we presented several architectures to perform multiple recognition tasks on laparoscopic videos. In this technical report, we present the tool presence detection results using two architectures: (1) a single-task architecture designed to perform solely the tool presence detection task and (2) a multi-task architecture designed to perform jointly phase recognition and tool presence detection. The results show that the multi-task network only slightly improves the tool presence detection results. In constrast, a significant improvement is obtained when there are more data available to train the networks. This significant improvement can be regarded as a call for action for other institutions to start working toward publishing more datasets into the community, so that better models could be generated to perform the task.

Andru P. Twinanda, Didier Mutter, Jacques Marescaux et al.

The surgical workflow challenge at M2CAI 2016 consists of identifying 8 surgical phases in cholecystectomy procedures. Here, we propose to use deep architectures that are based on our previous work where we presented several architectures to perform multiple recognition tasks on laparoscopic videos. In this technical report, we present the phase recognition results using two architectures: (1) a single-task architecture designed to perform solely the surgical phase recognition task and (2) a multi-task architecture designed to perform jointly phase recognition and tool presence detection. On top of these architectures we propose to use two different approaches to enforce the temporal constraints of the surgical workflow: (1) HMM-based and (2) LSTM-based pipelines. The results show that the LSTM-based approach is able to outperform the HMM-based approach and also to properly enforce the temporal constraints into the recognition process.

Abdolrahim Kadkhodamohammadi, Afshin Gangi, Michel de Mathelin et al.

Reliable human pose estimation (HPE) is essential to many clinical applications, such as surgical workflow analysis, radiation safety monitoring and human-robot cooperation. Proposed methods for the operating room (OR) rely either on foreground estimation using a multi-camera system, which is a challenge in real ORs due to color similarities and frequent illumination changes, or on wearable sensors or markers, which are invasive and therefore difficult to introduce in the room. Instead, we propose a novel approach based on Pictorial Structures (PS) and on RGB-D data, which can be easily deployed in real ORs. We extend the PS framework in two ways. First, we build robust and discriminative part detectors using both color and depth images. We also present a novel descriptor for depth images, called histogram of depth differences (HDD). Second, we extend PS to 3D by proposing 3D pairwise constraints and a new method that makes exact inference tractable. Our approach is evaluated for pose estimation and clinician detection on a challenging RGB-D dataset recorded in a busy operating room during live surgeries. We conduct series of experiments to study the different part detectors in conjunction with the various 2D or 3D pairwise constraints. Our comparisons demonstrate that 3D PS with RGB-D part detectors significantly improves the results in a visually challenging operating environment.

Andru P. Twinanda, Sherif Shehata, Didier Mutter et al.

Surgical workflow recognition has numerous potential medical applications, such as the automatic indexing of surgical video databases and the optimization of real-time operating room scheduling, among others. As a result, phase recognition has been studied in the context of several kinds of surgeries, such as cataract, neurological, and laparoscopic surgeries. In the literature, two types of features are typically used to perform this task: visual features and tool usage signals. However, the visual features used are mostly handcrafted. Furthermore, the tool usage signals are usually collected via a manual annotation process or by using additional equipment. In this paper, we propose a novel method for phase recognition that uses a convolutional neural network (CNN) to automatically learn features from cholecystectomy videos and that relies uniquely on visual information. In previous studies, it has been shown that the tool signals can provide valuable information in performing the phase recognition task. Thus, we present a novel CNN architecture, called EndoNet, that is designed to carry out the phase recognition and tool presence detection tasks in a multi-task manner. To the best of our knowledge, this is the first work proposing to use a CNN for multiple recognition tasks on laparoscopic videos. Extensive experimental comparisons to other methods show that EndoNet yields state-of-the-art results for both tasks.