External validation, radiological evaluation, and development of deep learning automatic lung segmentation in contrast-enhanced chest CT.

OBJECTIVES: There is a need for CT pulmonary angiography (CTPA) lung segmentation models. Clinical translation requires radiological evaluation of model outputs, understanding of limitations, and identification of failure points. This multicentre study aims to develop an accurate CTPA lung segmentation model, with evaluation of outputs in two diverse patient cohorts with pulmonary hypertension (PH) and interstitial lung disease (ILD). METHODS: This retrospective study develops an nnU-Net-based segmentation model using data from two specialist centres (UK and USA). Model was trained (n = 37), tested (n = 12), and clinically evaluated (n = 176) on a diverse 'real-world' cohort of 225 PH patients with volumetric CTPAs. Dice score coefficient (DSC) and normalised surface distance (NSD) were used for testing. Clinical evaluation of outputs was performed by two radiologists who assessed clinical significance of errors. External validation was performed on heterogenous contrast and non-contrast scans from 28 ILD patients. RESULTS: A total of 225 PH and 28 ILD patients with diverse demographic and clinical characteristics were evaluated. Mean accuracy, DSC, and NSD scores were 0.998 (95% CI 0.9976, 0.9989), 0.990 (0.9840, 0.9962), and 0.983 (0.9686, 0.9972) respectively. There were no segmentation failures. On radiological review, 82% and 71% of internal and external cases respectively had no errors. Eighteen percent and 25% respectively had clinically insignificant errors. Peripheral atelectasis and consolidation were common causes for suboptimal segmentation. One external case (0.5%) with patulous oesophagus had a clinically significant error. CONCLUSION: State-of-the-art CTPA lung segmentation model provides accurate outputs with minimal clinical errors on evaluation across two diverse cohorts with PH and ILD. CLINICAL RELEVANCE: Clinical translation of artificial intelligence models requires radiological review and understanding of model limitations. This study develops an externally validated state-of-the-art model with robust radiological review. Intended clinical use is in techniques such as lung volume or parenchymal disease quantification. KEY POINTS: • Accurate, externally validated CT pulmonary angiography (CTPA) lung segmentation model tested in two large heterogeneous clinical cohorts (pulmonary hypertension and interstitial lung disease). • No segmentation failures and robust review of model outputs by radiologists found 1 (0.5%) clinically significant segmentation error. • Intended clinical use of this model is a necessary step in techniques such as lung volume, parenchymal disease quantification, or pulmonary vessel analysis.

Fully automated breast segmentation on spiral breast computed tomography images.

INTRODUCTION: The quantification of the amount of the glandular tissue and breast density is important to assess breast cancer risk. Novel photon-counting breast computed tomography (CT) technology has the potential to quantify them. For accurate analysis, a dedicated method to segment the breast components-the adipose and glandular tissue, skin, pectoralis muscle, skinfold section, rib, and implant-is required. We propose a fully automated breast segmentation method for breast CT images. METHODS: The framework consists of four parts: (1) investigate, (2) segment the components excluding adipose and glandular tissue, (3) assess the breast density, and (4) iteratively segment the glandular tissue according to the estimated density. For the method, adapted seeded watershed and region growing algorithm were dedicatedly developed for the breast CT images and optimized on 68 breast images. The segmentation performance was qualitatively (five-point Likert scale) and quantitatively (Dice similarity coefficient [DSC] and difference coefficient [DC]) demonstrated according to human reading by experienced radiologists. RESULTS: The performance evaluation on each component and overall segmentation for 17 breast CT images resulted in DSCs ranging 0.90-0.97 and in DCs 0.01-0.08. The readers rated 4.5-4.8 (5 highest score) with an excellent inter-reader agreement. The breast density varied by 3.7%-7.1% when including mis-segmented muscle or skin. CONCLUSION: The automatic segmentation results coincided with the human expert's reading. The accurate segmentation is important to avoid the significant bias in breast density analysis. Our method enables accurate quantification of the breast density and amount of the glandular tissue that is directly related to breast cancer risk.

Large language models in radiology: fundamentals, applications, ethical considerations, risks, and future directions.

With the advent of large language models (LLMs), the artificial intelligence revolution in medicine and radiology is now more tangible than ever. Every day, an increasingly large number of articles are published that utilize LLMs in radiology. To adopt and safely implement this new technology in the field, radiologists should be familiar with its key concepts, understand at least the technical basics, and be aware of the potential risks and ethical considerations that come with it. In this review article, the authors provide an overview of the LLMs that might be relevant to the radiology community and include a brief discussion of their short history, technical basics, ChatGPT, prompt engineering, potential applications in medicine and radiology, advantages, disadvantages and risks, ethical and regulatory considerations, and future directions.

Bias in artificial intelligence for medical imaging: fundamentals, detection, avoidance, mitigation, challenges, ethics, and prospects.

Although artificial intelligence (AI) methods hold promise for medical imaging-based prediction tasks, their integration into medical practice may present a double-edged sword due to bias (i.e., systematic errors). AI algorithms have the potential to mitigate cognitive biases in human interpretation, but extensive research has highlighted the tendency of AI systems to internalize biases within their model. This fact, whether intentional or not, may ultimately lead to unintentional consequences in the clinical setting, potentially compromising patient outcomes. This concern is particularly important in medical imaging, where AI has been more progressively and widely embraced than any other medical field. A comprehensive understanding of bias at each stage of the AI pipeline is therefore essential to contribute to developing AI solutions that are not only less biased but also widely applicable. This international collaborative review effort aims to increase awareness within the medical imaging community about the importance of proactively identifying and addressing AI bias to prevent its negative consequences from being realized later. The authors began with the fundamentals of bias by explaining its different definitions and delineating various potential sources. Strategies for detecting and identifying bias were then outlined, followed by a review of techniques for its avoidance and mitigation. Moreover, ethical dimensions, challenges encountered, and prospects were discussed.

Adapted large language models can outperform medical experts in clinical text summarization.

Analyzing vast textual data and summarizing key information from electronic health records imposes a substantial burden on how clinicians allocate their time. Although large language models (LLMs) have shown promise in natural language processing (NLP) tasks, their effectiveness on a diverse range of clinical summarization tasks remains unproven. Here we applied adaptation methods to eight LLMs, spanning four distinct clinical summarization tasks: radiology reports, patient questions, progress notes and doctor-patient dialogue. Quantitative assessments with syntactic, semantic and conceptual NLP metrics reveal trade-offs between models and adaptation methods. A clinical reader study with 10 physicians evaluated summary completeness, correctness and conciseness; in most cases, summaries from our best-adapted LLMs were deemed either equivalent (45%) or superior (36%) compared with summaries from medical experts. The ensuing safety analysis highlights challenges faced by both LLMs and medical experts, as we connect errors to potential medical harm and categorize types of fabricated information. Our research provides evidence of LLMs outperforming medical experts in clinical text summarization across multiple tasks. This suggests that integrating LLMs into clinical workflows could alleviate documentation burden, allowing clinicians to focus more on patient care.

FDA-cleared artificial intelligence and machine learning-based medical devices and their 510(k) predicate networks.

The US Food and Drug Administration is clearing an increasing number of artificial intelligence and machine learning (AI/ML)-based medical devices through the 510(k) pathway. This pathway allows clearance if the device is substantially equivalent to a former cleared device (ie, predicate). We analysed the predicate networks of cleared AI/ML-based medical devices (cleared between 2019 and 2021), their underlying tasks, and recalls. More than a third of cleared AI/ML-based medical devices originated from non-AI/ML-based medical devices in the first generation. Devices with the longest time since the last predicate device with an AI/ML component were haematology (2001), radiology (2001), and cardiovascular devices (2008). Especially for devices in radiology, the AI/ML tasks changed frequently along the device's predicate network, raising safety concerns. To date, only a few recalls might have affected the AI/ML components. To improve patient care, a stronger focus should be placed on the distinctive characteristics of AI/ML when defining substantial equivalence between a new AI/ML-based medical device and predicate devices.

Clinical Text Summarization: Adapting Large Language Models Can Outperform Human Experts.

Sifting through vast textual data and summarizing key information from electronic health records (EHR) imposes a substantial burden on how clinicians allocate their time. Although large language models (LLMs) have shown immense promise in natural language processing (NLP) tasks, their efficacy on a diverse range of clinical summarization tasks has not yet been rigorously demonstrated. In this work, we apply domain adaptation methods to eight LLMs, spanning six datasets and four distinct clinical summarization tasks: radiology reports, patient questions, progress notes, and doctor-patient dialogue. Our thorough quantitative assessment reveals trade-offs between models and adaptation methods in addition to instances where recent advances in LLMs may not improve results. Further, in a clinical reader study with ten physicians, we show that summaries from our best-adapted LLMs are preferable to human summaries in terms of completeness and correctness. Our ensuing qualitative analysis highlights challenges faced by both LLMs and human experts. Lastly, we correlate traditional quantitative NLP metrics with reader study scores to enhance our understanding of how these metrics align with physician preferences. Our research marks the first evidence of LLMs outperforming human experts in clinical text summarization across multiple tasks. This implies that integrating LLMs into clinical workflows could alleviate documentation burden, empowering clinicians to focus more on personalized patient care and the inherently human aspects of medicine.