Beyond Multiple-Choice Accuracy: Real-World Challenges of Implementing Large Language Models in Healthcare

Large Language Models (LLMs) have gained significant attention in the medical domain for their human-level capabilities, leading to increased efforts to explore their potential in various healthcare applications. However, despite such a promising future, there are multiple challenges and obstacles that remain for their real-world uses in practical settings. This work discusses key challenges for LLMs in medical applications from four unique aspects: operational vulnerabilities, ethical and social considerations, performance and assessment difficulties, and legal and regulatory compliance. Addressing these challenges is crucial for leveraging LLMs to their full potential and ensuring their responsible integration into healthcare.

Demystifying Large Language Models for Medicine: A Primer

Large language models (LLMs) represent a transformative class of AI tools capable of revolutionizing various aspects of healthcare by generating human-like responses across diverse contexts and adapting to novel tasks following human instructions. Their potential application spans a broad range of medical tasks, such as clinical documentation, matching patients to clinical trials, and answering medical questions. In this primer paper, we propose an actionable guideline to help healthcare professionals more efficiently utilize LLMs in their work, along with a set of best practices. This approach consists of several main phases, including formulating the task, choosing LLMs, prompt engineering, fine-tuning, and deployment. We start with the discussion of critical considerations in identifying healthcare tasks that align with the core capabilities of LLMs and selecting models based on the selected task and data, performance requirements, and model interface. We then review the strategies, such as prompt engineering and fine-tuning, to adapt standard LLMs to specialized medical tasks. Deployment considerations, including regulatory compliance, ethical guidelines, and continuous monitoring for fairness and bias, are also discussed. By providing a structured step-by-step methodology, this tutorial aims to equip healthcare professionals with the tools necessary to effectively integrate LLMs into clinical practice, ensuring that these powerful technologies are applied in a safe, reliable, and impactful manner.

Deep Learning Segmentation of Ascites on Abdominal CT Scans for Automatic Volume Quantification

Purpose: To evaluate the performance of an automated deep learning method in detecting ascites and subsequently quantifying its volume in patients with liver cirrhosis and ovarian cancer. Materials and Methods: This retrospective study included contrast-enhanced and non-contrast abdominal-pelvic CT scans of patients with cirrhotic ascites and patients with ovarian cancer from two institutions, National Institutes of Health (NIH) and University of Wisconsin (UofW). The model, trained on The Cancer Genome Atlas Ovarian Cancer dataset (mean age, 60 years +/- 11 [s.d.]; 143 female), was tested on two internal (NIH-LC and NIH-OV) and one external dataset (UofW-LC). Its performance was measured by the Dice coefficient, standard deviations, and 95% confidence intervals, focusing on ascites volume in the peritoneal cavity. Results: On NIH-LC (25 patients; mean age, 59 years +/- 14 [s.d.]; 14 male) and NIH-OV (166 patients; mean age, 65 years +/- 9 [s.d.]; all female), the model achieved Dice scores of 0.855 +/- 0.061 (CI: 0.831-0.878) and 0.826 +/- 0.153 (CI: 0.764-0.887), with median volume estimation errors of 19.6% (IQR: 13.2-29.0) and 5.3% (IQR: 2.4-9.7) respectively. On UofW-LC (124 patients; mean age, 46 years +/- 12 [s.d.]; 73 female), the model had a Dice score of 0.830 +/- 0.107 (CI: 0.798-0.863) and median volume estimation error of 9.7% (IQR: 4.5-15.1). The model showed strong agreement with expert assessments, with r^2 values of 0.79, 0.98, and 0.97 across the test sets. Conclusion: The proposed deep learning method performed well in segmenting and quantifying the volume of ascites in concordance with expert radiologist assessments.

Shadow and Light: Digitally Reconstructed Radiographs for Disease Classification

In this paper, we introduce DRR-RATE, a large-scale synthetic chest X-ray dataset derived from the recently released CT-RATE dataset. DRR-RATE comprises of 50,188 frontal Digitally Reconstructed Radiographs (DRRs) from 21,304 unique patients. Each image is paired with a corresponding radiology text report and binary labels for 18 pathology classes. Given the controllable nature of DRR generation, it facilitates the inclusion of lateral view images and images from any desired viewing position. This opens up avenues for research into new and novel multimodal applications involving paired CT, X-ray images from various views, text, and binary labels. We demonstrate the applicability of DRR-RATE alongside existing large-scale chest X-ray resources, notably the CheXpert dataset and CheXnet model. Experiments demonstrate that CheXnet, when trained and tested on the DRR-RATE dataset, achieves sufficient to high AUC scores for the six common pathologies cited in common literature: Atelectasis, Cardiomegaly, Consolidation, Lung Lesion, Lung Opacity, and Pleural Effusion. Additionally, CheXnet trained on the CheXpert dataset can accurately identify several pathologies, even when operating out of distribution. This confirms that the generated DRR images effectively capture the essential pathology features from CT images. The dataset and labels are publicly accessible at https://huggingface.co/datasets/farrell236/DRR-RATE.

MRISegmentator-Abdomen: A Fully Automated Multi-Organ and Structure Segmentation Tool for T1-weighted Abdominal MRI

Background: Segmentation of organs and structures in abdominal MRI is useful for many clinical applications, such as disease diagnosis and radiotherapy. Current approaches have focused on delineating a limited set of abdominal structures (13 types). To date, there is no publicly available abdominal MRI dataset with voxel-level annotations of multiple organs and structures. Consequently, a segmentation tool for multi-structure segmentation is also unavailable. Methods: We curated a T1-weighted abdominal MRI dataset consisting of 195 patients who underwent imaging at National Institutes of Health (NIH) Clinical Center. The dataset comprises of axial pre-contrast T1, arterial, venous, and delayed phases for each patient, thereby amounting to a total of 780 series (69,248 2D slices). Each series contains voxel-level annotations of 62 abdominal organs and structures. A 3D nnUNet model, dubbed as MRISegmentator-Abdomen (MRISegmentator in short), was trained on this dataset, and evaluation was conducted on an internal test set and two large external datasets: AMOS22 and Duke Liver. The predicted segmentations were compared against the ground-truth using the Dice Similarity Coefficient (DSC) and Normalized Surface Distance (NSD). Findings: MRISegmentator achieved an average DSC of 0.861$\pm$0.170 and a NSD of 0.924$\pm$0.163 in the internal test set. On the AMOS22 dataset, MRISegmentator attained an average DSC of 0.829$\pm$0.133 and a NSD of 0.908$\pm$0.067. For the Duke Liver dataset, an average DSC of 0.933$\pm$0.015 and a NSD of 0.929$\pm$0.021 was obtained. Interpretation: The proposed MRISegmentator provides automatic, accurate, and robust segmentations of 62 organs and structures in T1-weighted abdominal MRI sequences. The tool has the potential to accelerate research on various clinical topics, such as abnormality detection, radiotherapy, disease classification among others.

Decomposing Vision-based LLM Predictions for Auto-Evaluation with GPT-4

The volume of CT exams being done in the world has been rising every year, which has led to radiologist burn-out. Large Language Models (LLMs) have the potential to reduce their burden, but their adoption in the clinic depends on radiologist trust, and easy evaluation of generated content. Presently, many automated methods are available to evaluate the reports generated for chest radiographs, but such an approach is not available for CT presently. In this paper, we propose a novel evaluation framework to judge the capabilities of vision-language LLMs in generating accurate summaries of CT-based abnormalities. CT slices containing an abnormality (e.g., lesion) were input to a vision-based LLM (GPT-4V, LLaVA-Med, and RadFM), and it generated a free-text summary of the predicted characteristics of the abnormality. Next, a GPT-4 model decomposed the summary into specific aspects (body part, location, type, and attributes), automatically evaluated the characteristics against the ground-truth, and generated a score for each aspect based on its clinical relevance and factual accuracy. These scores were then contrasted against those obtained from a clinician, and a high correlation ( 85%, p < .001) was observed. Although GPT-4V outperformed other models in our evaluation, it still requires overall improvement. Our evaluation method offers valuable insights into the specific areas that need the most enhancement, guiding future development in this field.

Leveraging Professional Radiologists' Expertise to Enhance LLMs' Evaluation for Radiology Reports

In radiology, Artificial Intelligence (AI) has significantly advanced report generation, but automatic evaluation of these AI-produced reports remains challenging. Current metrics, such as Conventional Natural Language Generation (NLG) and Clinical Efficacy (CE), often fall short in capturing the semantic intricacies of clinical contexts or overemphasize clinical details, undermining report clarity. To overcome these issues, our proposed method synergizes the expertise of professional radiologists with Large Language Models (LLMs), like GPT-3.5 and GPT-4 1. Utilizing In-Context Instruction Learning (ICIL) and Chain of Thought (CoT) reasoning, our approach aligns LLM evaluations with radiologist standards, enabling detailed comparisons between human and AI generated reports. This is further enhanced by a Regression model that aggregates sentence evaluation scores. Experimental results show that our "Detailed GPT-4 (5-shot)" model achieves a 0.48 score, outperforming the METEOR metric by 0.19, while our "Regressed GPT-4" model shows even greater alignment with expert evaluations, exceeding the best existing metric by a 0.35 margin. Moreover, the robustness of our explanations has been validated through a thorough iterative strategy. We plan to publicly release annotations from radiology experts, setting a new standard for accuracy in future assessments. This underscores the potential of our approach in enhancing the quality assessment of AI-driven medical reports.

Enhanced Muscle and Fat Segmentation for CT-Based Body Composition Analysis: A Comparative Study

Purpose: Body composition measurements from routine abdominal CT can yield personalized risk assessments for asymptomatic and diseased patients. In particular, attenuation and volume measures of muscle and fat are associated with important clinical outcomes, such as cardiovascular events, fractures, and death. This study evaluates the reliability of an Internal tool for the segmentation of muscle and fat (subcutaneous and visceral) as compared to the well-established public TotalSegmentator tool. Methods: We assessed the tools across 900 CT series from the publicly available SAROS dataset, focusing on muscle, subcutaneous fat, and visceral fat. The Dice score was employed to assess accuracy in subcutaneous fat and muscle segmentation. Due to the lack of ground truth segmentations for visceral fat, Cohen's Kappa was utilized to assess segmentation agreement between the tools. Results: Our Internal tool achieved a 3% higher Dice (83.8 vs. 80.8) for subcutaneous fat and a 5% improvement (87.6 vs. 83.2) for muscle segmentation respectively. A Wilcoxon signed-rank test revealed that our results were statistically different with p<0.01. For visceral fat, the Cohen's kappa score of 0.856 indicated near-perfect agreement between the two tools. Our internal tool also showed very strong correlations for muscle volume (R^2=0.99), muscle attenuation (R^2=0.93), and subcutaneous fat volume (R^2=0.99) with a moderate correlation for subcutaneous fat attenuation (R^2=0.45). Conclusion: Our findings indicated that our Internal tool outperformed TotalSegmentator in measuring subcutaneous fat and muscle. The high Cohen's Kappa score for visceral fat suggests a reliable level of agreement between the two tools. These results demonstrate the potential of our tool in advancing the accuracy of body composition analysis.

Semantic Image Synthesis for Abdominal CT

As a new emerging and promising type of generative models, diffusion models have proven to outperform Generative Adversarial Networks (GANs) in multiple tasks, including image synthesis. In this work, we explore semantic image synthesis for abdominal CT using conditional diffusion models, which can be used for downstream applications such as data augmentation. We systematically evaluated the performance of three diffusion models, as well as to other state-of-the-art GAN-based approaches, and studied the different conditioning scenarios for the semantic mask. Experimental results demonstrated that diffusion models were able to synthesize abdominal CT images with better quality. Additionally, encoding the mask and the input separately is more effective than na\"ive concatenating.

High-Fidelity Image Synthesis from Pulmonary Nodule Lesion Maps using Semantic Diffusion Model

Lung cancer has been one of the leading causes of cancer-related deaths worldwide for years. With the emergence of deep learning, computer-assisted diagnosis (CAD) models based on learning algorithms can accelerate the nodule screening process, providing valuable assistance to radiologists in their daily clinical workflows. However, developing such robust and accurate models often requires large-scale and diverse medical datasets with high-quality annotations. Generating synthetic data provides a pathway for augmenting datasets at a larger scale. Therefore, in this paper, we explore the use of Semantic Diffusion Mod- els (SDM) to generate high-fidelity pulmonary CT images from segmentation maps. We utilize annotation information from the LUNA16 dataset to create paired CT images and masks, and assess the quality of the generated images using the Frechet Inception Distance (FID), as well as on two common clinical downstream tasks: nodule detection and nodule localization. Achieving improvements of 3.96% for detection accuracy and 8.50% for AP50 in nodule localization task, respectively, demonstrates the feasibility of the approach.