Errol Colak

Kathryn Wantlin, Chenwei Wu, Shih-Cheng Huang et al. · stanford

Medical data poses a daunting challenge for AI algorithms: it exists in many different modalities, experiences frequent distribution shifts, and suffers from a scarcity of examples and labels. Recent advances, including transformers and self-supervised learning, promise a more universal approach that can be applied flexibly across these diverse conditions. To measure and drive progress in this direction, we present BenchMD: a benchmark that tests how well unified, modality-agnostic methods, including architectures and training techniques (e.g. self-supervised learning, ImageNet pretraining),perform on a diverse array of clinically-relevant medical tasks. BenchMD combines 19 publicly available datasets for 7 medical modalities, including 1D sensor data, 2D images, and 3D volumetric scans. Our benchmark reflects real-world data constraints by evaluating methods across a range of dataset sizes, including challenging few-shot settings that incentivize the use of pretraining. Finally, we evaluate performance on out-of-distribution data collected at different hospitals than the training data, representing naturally-occurring distribution shifts that frequently degrade the performance of medical AI models. Our baseline results demonstrate that no unified learning technique achieves strong performance across all modalities, leaving ample room for improvement on the benchmark. Code is released at https://github.com/rajpurkarlab/BenchMD.

Adam E. Flanders, Yifan Peng, Luciano Prevedello et al.

Purpose: The purpose of this study was to determine if an ensemble of multiple LLM agents could be used collectively to provide a more reliable assessment of a pixel-based AI triage tool than a single LLM. Methods: 29,766 non-contrast CT head exams from fourteen hospitals were processed by a commercial intracranial hemorrhage (ICH) AI detection tool. Radiology reports were analyzed by an ensemble of eight open-source LLM models and a HIPAA compliant internal version of GPT-4o using a single multi-shot prompt that assessed for presence of ICH. 1,726 examples were manually reviewed. Performance characteristics of the eight open-source models and consensus were compared to GPT-4o. Three ideal consensus LLM ensembles were tested for rating the performance of the triage tool. Results: The cohort consisted of 29,766 head CTs exam-report pairs. The highest AUC performance was achieved with llama3.3:70b and GPT-4o (AUC= 0.78). The average precision was highest for Llama3.3:70b and GPT-4o (AP=0.75 & 0.76). Llama3.3:70b had the highest F1 score (0.81) and recall (0.85), greater precision (0.78), specificity (0.72), and MCC (0.57). Using MCC (95% CI) the ideal combination of LLMs were: Full-9 Ensemble 0.571 (0.552-0.591), Top-3 Ensemble 0.558 (0.537-0.579), Consensus 0.556 (0.539-0.574), and GPT4o 0.522 (0.500-0.543). No statistically significant differences were observed between Top-3, Full-9, and Consensus (p > 0.05). Conclusion: An ensemble of medium to large sized open-source LLMs provides a more consistent and reliable method to derive a ground truth retrospective evaluation of a clinical AI triage tool over a single LLM alone.

Edward H. Lee, Brendan Kelly, Emre Altinmakas et al.

While it is well known that population differences from genetics, sex, race, and environmental factors contribute to disease, AI studies in medicine have largely focused on locoregional patient cohorts with less diverse data sources. Such limitation stems from barriers to large-scale data share and ethical concerns over data privacy. Federated learning (FL) is one potential pathway for AI development that enables learning across hospitals without data share. In this study, we show the results of various FL strategies on one of the largest and most diverse COVID-19 chest CT datasets: 21 participating hospitals across five continents that comprise >10,000 patients with >1 million images. We also propose an FL strategy that leverages synthetically generated data to overcome class and size imbalances. We also describe the sources of data heterogeneity in the context of FL, and show how even among the correctly labeled populations, disparities can arise due to these biases.

Yishu Wei, Adam E. Flanders, Errol Colak et al.

Multimodal large language models have demonstrated comparable performance to that of radiology trainees on multiple-choice board-style exams. However, to develop clinically useful multimodal LLM tools, high-quality benchmarks curated by domain experts are essential. To curate released and holdout datasets of 100 chest radiographic studies each and propose an artificial intelligence (AI)-assisted expert labeling procedure to allow radiologists to label studies more efficiently. A total of 13,735 deidentified chest radiographs and their corresponding reports from the MIDRC were used. GPT-4o extracted abnormal findings from the reports, which were then mapped to 12 benchmark labels with a locally hosted LLM (Phi-4-Reasoning). From these studies, 1,000 were sampled on the basis of the AI-suggested benchmark labels for expert review; the sampling algorithm ensured that the selected studies were clinically relevant and captured a range of difficulty levels. Seventeen chest radiologists participated, and they marked "Agree all", "Agree mostly" or "Disagree" to indicate their assessment of the correctness of the LLM suggested labels. Each chest radiograph was evaluated by three experts. Of these, at least two radiologists selected "Agree All" for 381 radiographs. From this set, 200 were selected, prioritizing those with less common or multiple finding labels, and divided into 100 released radiographs and 100 reserved as the holdout dataset. The holdout dataset is used exclusively by RSNA to independently evaluate different models. A benchmark of 200 chest radiographic studies with 12 benchmark labels was created and made publicly available https://imaging.rsna.org, with each chest radiograph verified by three radiologists. In addition, an AI-assisted labeling procedure was developed to help radiologists label at scale, minimize unnecessary omissions, and support a semicollaborative environment.

Tyler J. Richards, Adam E. Flanders, Errol Colak et al.

The Radiological Society of North America (RSNA) Lumbar Degenerative Imaging Spine Classification (LumbarDISC) dataset is the largest publicly available dataset of adult MRI lumbar spine examinations annotated for degenerative changes. The dataset includes 2,697 patients with a total of 8,593 image series from 8 institutions across 6 countries and 5 continents. The dataset is available for free for non-commercial use via Kaggle and RSNA Medical Imaging Resource of AI (MIRA). The dataset was created for the RSNA 2024 Lumbar Spine Degenerative Classification competition where competitors developed deep learning models to grade degenerative changes in the lumbar spine. The degree of spinal canal, subarticular recess, and neural foraminal stenosis was graded at each intervertebral disc level in the lumbar spine. The images were annotated by expert volunteer neuroradiologists and musculoskeletal radiologists from the RSNA, American Society of Neuroradiology, and the American Society of Spine Radiology. This dataset aims to facilitate research and development in machine learning and lumbar spine imaging to lead to improved patient care and clinical efficiency.

Florian Kofler, Felix Meissen, Felix Steinbauer et al.

A myriad of algorithms for the automatic analysis of brain MR images is available to support clinicians in their decision-making. For brain tumor patients, the image acquisition time series typically starts with an already pathological scan. This poses problems, as many algorithms are designed to analyze healthy brains and provide no guarantee for images featuring lesions. Examples include, but are not limited to, algorithms for brain anatomy parcellation, tissue segmentation, and brain extraction. To solve this dilemma, we introduce the BraTS inpainting challenge. Here, the participants explore inpainting techniques to synthesize healthy brain scans from lesioned ones. The following manuscript contains the task formulation, dataset, and submission procedure. Later, it will be updated to summarize the findings of the challenge. The challenge is organized as part of the ASNR-BraTS MICCAI challenge.

Edward H. Lee, Mario Michael Krell, Alexander Tsyplikhin et al.

Differentially private SGD (DPSGD) has recently shown promise in deep learning. However, compared to non-private SGD, the DPSGD algorithm places computational overheads that can undo the benefit of batching in GPUs. Micro-batching is a common method to alleviate this and is fully supported in the TensorFlow Privacy library (TFDP). However, it degrades accuracy. We propose NanoBatch Privacy, a lightweight add-on to TFDP to be used on Graphcore IPUs by leveraging batch size of 1 (without microbatching) and gradient accumulation. This allows us to achieve large total batch sizes with minimal impacts to throughput. Second, we illustrate using Cifar-10 how larger batch sizes are not necessarily optimal from a privacy versus utility perspective. On ImageNet, we achieve more than 15x speedup over TFDP versus 8x A100s and significant speedups even across libraries such as Opacus. We also provide two extensions: 1) DPSGD for pipelined models and 2) per-layer clipping that is 15x faster than the Opacus implementation on 8x A100s. Finally as an application case study, we apply NanoBatch training for use on private Covid-19 chest CT prediction.

Ujjwal Baid, Satyam Ghodasara, Suyash Mohan et al.

The BraTS 2021 challenge celebrates its 10th anniversary and is jointly organized by the Radiological Society of North America (RSNA), the American Society of Neuroradiology (ASNR), and the Medical Image Computing and Computer Assisted Interventions (MICCAI) society. Since its inception, BraTS has been focusing on being a common benchmarking venue for brain glioma segmentation algorithms, with well-curated multi-institutional multi-parametric magnetic resonance imaging (mpMRI) data. Gliomas are the most common primary malignancies of the central nervous system, with varying degrees of aggressiveness and prognosis. The RSNA-ASNR-MICCAI BraTS 2021 challenge targets the evaluation of computational algorithms assessing the same tumor compartmentalization, as well as the underlying tumor's molecular characterization, in pre-operative baseline mpMRI data from 2,040 patients. Specifically, the two tasks that BraTS 2021 focuses on are: a) the segmentation of the histologically distinct brain tumor sub-regions, and b) the classification of the tumor's O[6]-methylguanine-DNA methyltransferase (MGMT) promoter methylation status. The performance evaluation of all participating algorithms in BraTS 2021 will be conducted through the Sage Bionetworks Synapse platform (Task 1) and Kaggle (Task 2), concluding in distributing to the top ranked participants monetary awards of $60,000 collectively.

Hojjat Salehinejad, Jumpei Kitamura, Noah Ditkofsky et al.

Machine learning (ML) holds great promise in transforming healthcare. While published studies have shown the utility of ML models in interpreting medical imaging examinations, these are often evaluated under laboratory settings. The importance of real world evaluation is best illustrated by case studies that have documented successes and failures in the translation of these models into clinical environments. A key prerequisite for the clinical adoption of these technologies is demonstrating generalizable ML model performance under real world circumstances. The purpose of this study was to demonstrate that ML model generalizability is achievable in medical imaging with the detection of intracranial hemorrhage (ICH) on non-contrast computed tomography (CT) scans serving as the use case. An ML model was trained using 21,784 scans from the RSNA Intracranial Hemorrhage CT dataset while generalizability was evaluated using an external validation dataset obtained from our busy trauma and neurosurgical center. This real world external validation dataset consisted of every unenhanced head CT scan (n = 5,965) performed in our emergency department in 2019 without exclusion. The model demonstrated an AUC of 98.4%, sensitivity of 98.8%, and specificity of 98.0%, on the test dataset. On external validation, the model demonstrated an AUC of 95.4%, sensitivity of 91.3%, and specificity of 94.1%. Evaluating the ML model using a real world external validation dataset that is temporally and geographically distinct from the training dataset indicates that ML generalizability is achievable in medical imaging applications.

Hojjat Salehinejad, Edward Ho, Hui-Ming Lin et al.

Fractures of the cervical spine are a medical emergency and may lead to permanent paralysis and even death. Accurate diagnosis in patients with suspected fractures by computed tomography (CT) is critical to patient management. In this paper, we propose a deep convolutional neural network (DCNN) with a bidirectional long-short term memory (BLSTM) layer for the automated detection of cervical spine fractures in CT axial images. We used an annotated dataset of 3,666 CT scans (729 positive and 2,937 negative cases) to train and validate the model. The validation results show a classification accuracy of 70.92% and 79.18% on the balanced (104 positive and 104 negative cases) and imbalanced (104 positive and 419 negative cases) test datasets, respectively.

Hojjat Salehinejad, Sumeya Naqvi, Errol Colak et al.

In this paper, we propose a novel technique for sampling sequential images using a cylindrical transform in a cylindrical coordinate system for kidney semantic segmentation in abdominal computed tomography (CT). The images generated from a cylindrical transform augment a limited annotated set of images in three dimensions. This approach enables us to train contemporary classification deep convolutional neural networks (DCNNs) instead of fully convolutional networks (FCNs) for semantic segmentation. Typical semantic segmentation models segment a sequential set of images (e.g. CT or video) by segmenting each image independently. However, the proposed method not only considers the spatial dependency in the x-y plane, but also the spatial sequential dependency along the z-axis. The results show that classification DCNNs, trained on cylindrical transformed images, can achieve a higher segmentation performance value than FCNs using a limited number of annotated images.

Hojjat Salehinejad, Sharan Sankar, Joseph Barfett et al.

Recurrent neural networks (RNNs) are capable of learning features and long term dependencies from sequential and time-series data. The RNNs have a stack of non-linear units where at least one connection between units forms a directed cycle. A well-trained RNN can model any dynamical system; however, training RNNs is mostly plagued by issues in learning long-term dependencies. In this paper, we present a survey on RNNs and several new advances for newcomers and professionals in the field. The fundamentals and recent advances are explained and the research challenges are introduced.

Hojjat Salehinejad, Shahrokh Valaee, Tim Dowdell et al.

Medical datasets are often highly imbalanced with over-representation of common medical problems and a paucity of data from rare conditions. We propose simulation of pathology in images to overcome the above limitations. Using chest X-rays as a model medical image, we implement a generative adversarial network (GAN) to create artificial images based upon a modest sized labeled dataset. We employ a combination of real and artificial images to train a deep convolutional neural network (DCNN) to detect pathology across five classes of chest X-rays. Furthermore, we demonstrate that augmenting the original imbalanced dataset with GAN generated images improves performance of chest pathology classification using the proposed DCNN in comparison to the same DCNN trained with the original dataset alone. This improved performance is largely attributed to balancing of the dataset using GAN generated images, where image classes that are lacking in example images are preferentially augmented.

Hojjat Salehinejad, Shahrokh Valaee, Aren Mnatzakanian et al.

Radiology reports are an important means of communication between radiologists and other physicians. These reports express a radiologist's interpretation of a medical imaging examination and are critical in establishing a diagnosis and formulating a treatment plan. In this paper, we propose a Bi-directional convolutional neural network (Bi-CNN) model for the interpretation and classification of mammograms based on breast density and chest radiographic radiology reports based on the basis of chest pathology. The proposed approach helps to organize databases of radiology reports, retrieve them expeditiously, and evaluate the radiology report that could be used in an auditing system to decrease incorrect diagnoses. Our study revealed that the proposed Bi-CNN outperforms the random forest and the support vector machine methods.

Hojjat Salehinejad, Joseph Barfett, Parham Aarabi et al.

Pathfinding in hospitals is challenging for patients, visitors, and even employees. Many people have experienced getting lost due to lack of clear guidance, large footprint of hospitals, and confusing array of hospital wings. In this paper, we propose Halo; An indoor navigation application based on voice-user interaction to help provide directions for users without assistance of a localization system. The main challenge is accurate detection of origin and destination search terms. A custom convolutional neural network (CNN) is proposed to detect origin and destination search terms from transcription of a submitted speech query. The CNN is trained based on a set of queries tailored specifically for hospital and clinic environments. Performance of the proposed model is studied and compared with Levenshtein distance-based word matching.