Practical Evaluation of ChatGPT Performance for Radiology Report Generation.

RATIONALE AND OBJECTIVES: The process of generating radiology reports is often time-consuming and labor-intensive, prone to incompleteness, heterogeneity, and errors. By employing natural language processing (NLP)-based techniques, this study explores the potential for enhancing the efficiency of radiology report generation through the remarkable capabilities of ChatGPT (Generative Pre-training Transformer), a prominent large language model (LLM). MATERIALS AND METHODS: Using a sample of 1000 records from the Medical Information Mart for Intensive Care (MIMIC) Chest X-ray Database, this investigation employed Claude.ai to extract initial radiological report keywords. ChatGPT then generated radiology reports using a consistent 3-step prompt template outline. Various lexical and sentence similarity techniques were employed to evaluate the correspondence between the AI assistant-generated reports and reference reports authored by medical professionals. RESULTS: Results showed varying performance among NLP models, with Bart (Bidirectional and Auto-Regressive Transformers) and XLM (Cross-lingual Language Model) displaying high proficiency (mean similarity scores up to 99.3%), closely mirroring physician reports. Conversely, DeBERTa (Decoding-enhanced BERT with disentangled attention) and sequence-matching models scored lower, indicating less alignment with medical language. In the Impression section, the Word-Embedding model excelled with a mean similarity of 84.4%, while others like the Jaccard index showed lower performance. CONCLUSION: Overall, the study highlights significant variations across NLP models in their ability to generate radiology reports consistent with medical professionals' language. Pairwise comparisons and Kruskal-Wallis tests confirmed these differences, emphasizing the need for careful selection and evaluation of NLP models in radiology report generation. This research underscores the potential of ChatGPT to streamline and improve the radiology reporting process, with implications for enhancing efficiency and accuracy in clinical practice.

Efficient Clinical Information Extraction from Breast Radiology Reports in French.

Radiology reports contain crucial patient information, in addition to images, that can be automatically extracted for secondary uses such as clinical support and research for diagnosis. We tested several classifiers to classify 1,218 breast MRI reports in French from two Swiss clinical centers. Logistic regression performed better for both internal (accuracy > 0.95 and macro-F1 > 0.86) and external data (accuracy > 0.81 and macro-F1 > 0.41). Automating this task will facilitate efficient extraction of targeted clinical parameters and provide a good basis for future annotation processes through automatic pre-annotation.

Estimating the Severity of Oral Lesions Via Analysis of Cone Beam Computed Tomography Reports: A Proposed Deep Learning Model.

OBJECTIVES: Several factors such as unavailability of specialists, dental phobia, and financial difficulties may lead to a delay between receiving an oral radiology report and consulting a dentist. The primary aim of this study was to distinguish between high-risk and low-risk oral lesions according to the radiologist's reports of cone beam computed tomography (CBCT) images. Such a facility may be employed by dentist or his/her assistant to make the patient aware of the severity and the grade of the oral lesion and referral for immediate treatment or other follow-up care. METHODS: A total number of 1134 CBCT radiography reports owned by Shiraz University of Medical Sciences were collected. The severity level of each sample was specified by three experts, and an annotation was carried out accordingly. After preprocessing the data, a deep learning model, referred to as CNN-LSTM, was developed, which aims to detect the degree of severity of the problem based on analysis of the radiologist's report. Unlike traditional models which usually use a simple collection of words, the proposed deep model uses words embedded in dense vector representations, which empowers it to effectively capture semantic similarities. RESULTS: The results indicated that the proposed model outperformed its counterparts in terms of precision, recall, and F1 criteria. This suggests its potential as a reliable tool for early estimation of the severity of oral lesions. CONCLUSIONS: This study shows the effectiveness of deep learning in the analysis of textual reports and accurately distinguishing between high-risk and low-risk lesions. Employing the proposed model which can Provide timely warnings about the need for follow-up and prompt treatment can shield the patient from the risks associated with delays. CLINICAL SIGNIFICANCE: Our collaboratively collected and expert-annotated dataset serves as a valuable resource for exploratory research. The results demonstrate the pivotal role of our deep learning model could play in assessing the severity of oral lesions in dental reports.

From vision to text: A comprehensive review of natural image captioning in medical diagnosis and radiology report generation.

Natural Image Captioning (NIC) is an interdisciplinary research area that lies within the intersection of Computer Vision (CV) and Natural Language Processing (NLP). Several works have been presented on the subject, ranging from the early template-based approaches to the more recent deep learning-based methods. This paper conducts a survey in the area of NIC, especially focusing on its applications for Medical Image Captioning (MIC) and Diagnostic Captioning (DC) in the field of radiology. A review of the state-of-the-art is conducted summarizing key research works in NIC and DC to provide a wide overview on the subject. These works include existing NIC and MIC models, datasets, evaluation metrics, and previous reviews in the specialized literature. The revised work is thoroughly analyzed and discussed, highlighting the limitations of existing approaches and their potential implications in real clinical practice. Similarly, future potential research lines are outlined on the basis of the detected limitations.

Automatic text classification of prostate cancer malignancy scores in radiology reports using NLP models.

This paper presents the implementation of two automated text classification systems for prostate cancer findings based on the PI-RADS criteria. Specifically, a traditional machine learning model using XGBoost and a language model-based approach using RoBERTa were employed. The study focused on Spanish-language radiological MRI prostate reports, which has not been explored before. The results demonstrate that the RoBERTa model outperforms the XGBoost model, although both achieve promising results. Furthermore, the best-performing system was integrated into the radiological company's information systems as an API, operating in a real-world environment.

Visual Recalibration and Gating Enhancement Network for Radiology Report Generation.

Automatic radiology medical report generation is a necessary development of artificial intelligence technology in the health care. This technology serves to aid doctors in producing comprehensive diagnostic reports, alleviating the burdensome workloads of medical professionals. However, there are some challenges in generating radiological reports: (1) visual and textual data biases and (2) long-distance dependency problem. To tackle these issues, we design a visual recalibration and gating enhancement network (VRGE), which composes of the visual recalibration module and the gating enhancement module (gating enhancement module, GEM). Specifically, the visual recalibration module enhances the recognition of abnormal features in lesion areas of medical images. The GEM dynamically adjusts the contextual information in the report by introducing gating mechanisms, focusing on capturing professional medical terminology in medical text reports. We have conducted sufficient experiments on the public datasets of IU X-Ray to illustrate that the VRGE outperforms existing models.

Improving Radiology Report Generation Quality and Diversity through Reinforcement Learning and Text Augmentation.

Deep learning is revolutionizing radiology report generation (RRG) with the adoption of vision encoder-decoder (VED) frameworks, which transform radiographs into detailed medical reports. Traditional methods, however, often generate reports of limited diversity and struggle with generalization. Our research introduces reinforcement learning and text augmentation to tackle these issues, significantly improving report quality and variability. By employing RadGraph as a reward metric and innovating in text augmentation, we surpass existing benchmarks like BLEU4, ROUGE-L, F1CheXbert, and RadGraph, setting new standards for report accuracy and diversity on MIMIC-CXR and Open-i datasets. Our VED model achieves F1-scores of 66.2 for CheXbert and 37.8 for RadGraph on the MIMIC-CXR dataset, and 54.7 and 45.6, respectively, on Open-i. These outcomes represent a significant breakthrough in the RRG field. The findings and implementation of the proposed approach, aimed at enhancing diagnostic precision and radiological interpretations in clinical settings, are publicly available on GitHub to encourage further advancements in the field.

Evaluation of large language models performance against humans for summarizing MRI knee radiology reports: A feasibility study.

OBJECTIVES: This study addresses the critical need for accurate summarization in radiology by comparing various Large Language Model (LLM)-based approaches for automatic summary generation. With the increasing volume of patient information, accurately and concisely conveying radiological findings becomes crucial for effective clinical decision-making. Minor inaccuracies in summaries can lead to significant consequences, highlighting the need for reliable automated summarization tools. METHODS: We employed two language models - Text-to-Text Transfer Transformer (T5) and Bidirectional and Auto-Regressive Transformers (BART) - in both fine-tuned and zero-shot learning scenarios and compared them with a Recurrent Neural Network (RNN). Additionally, we conducted a comparative analysis of 100 MRI report summaries, using expert human judgment and criteria such as coherence, relevance, fluency, and consistency, to evaluate the models against the original radiologist summaries. To facilitate this, we compiled a dataset of 15,508 retrospective knee Magnetic Resonance Imaging (MRI) reports from our Radiology Information System (RIS), focusing on the findings section to predict the radiologist's summary. RESULTS: The fine-tuned models outperform the neural network and show superior performance in the zero-shot variant. Specifically, the T5 model achieved a Rouge-L score of 0.638. Based on the radiologist readers' study, the summaries produced by this model were found to be very similar to those produced by a radiologist, with about 70% similarity in fluency and consistency between the T5-generated summaries and the original ones. CONCLUSIONS: Technological advances, especially in NLP and LLM, hold great promise for improving and streamlining the summarization of radiological findings, thus providing valuable assistance to radiologists in their work.

Even with ChatGPT, race matters.

BACKGROUND: Applications of large language models such as ChatGPT are increasingly being studied. Before these technologies become entrenched, it is crucial to analyze whether they perpetuate racial inequities. METHODS: We asked Open AI's ChatGPT-3.5 and ChatGPT-4 to simplify 750 radiology reports with the prompt "I am a ___ patient. Simplify this radiology report:" while providing the context of the five major racial classifications on the U.S. census: White, Black or African American, American Indian or Alaska Native, Asian, and Native Hawaiian or other Pacific Islander. To ensure an unbiased analysis, the readability scores of the outputs were calculated and compared. RESULTS: Statistically significant differences were found in both models based on the racial context. For ChatGPT-3.5, output for White and Asian was at a significantly higher reading grade level than both Black or African American and American Indian or Alaska Native, among other differences. For ChatGPT-4, output for Asian was at a significantly higher reading grade level than American Indian or Alaska Native and Native Hawaiian or other Pacific Islander, among other differences. CONCLUSION: Here, we tested an application where we would expect no differences in output based on racial classification. Hence, the differences found are alarming and demonstrate that the medical community must remain vigilant to ensure large language models do not provide biased or otherwise harmful outputs.

Degree of Uncertainty in Reporting Imaging Findings for Necrotizing Enterocolitis: A Secondary Analysis from a Pilot Randomized Diagnostic Trial.

Diagnosis of necrotizing enterocolitis (NEC) relies heavily on imaging, but uncertainty in the language used in imaging reports can result in ambiguity, miscommunication, and potential diagnostic errors. To determine the degree of uncertainty in reporting imaging findings for NEC, we conducted a secondary analysis of the data from a previously completed pilot diagnostic randomized controlled trial (2019-2020). The study population comprised sixteen preterm infants with suspected NEC randomized to abdominal radiographs (AXRs) or AXR + bowel ultrasound (BUS). The level of uncertainty was determined using a four-point Likert scale. Overall, we reviewed radiology reports of 113 AXR and 24 BUS from sixteen preterm infants with NEC concern. The BUS reports showed less uncertainty for reporting pneumatosis, portal venous gas, and free air compared to AXR reports (pneumatosis: 1 [1-1.75) vs. 3 [2-3], p < 0.0001; portal venous gas: 1 [1-1] vs. 1 [1-1], p = 0.02; free air: 1 [1-1] vs. 2 [1-3], p < 0.0001). In conclusion, we found that BUS reports have a lower degree of uncertainty in reporting imaging findings of NEC compared to AXR reports. Whether the lower degree of uncertainty of BUS reports positively impacts clinical decision making in infants with possible NEC remains unknown.

Comparison of natural language processing algorithms in assessing the importance of head computed tomography reports written in Japanese.

PURPOSE: To propose a five-point scale for radiology report importance called Report Importance Category (RIC) and to compare the performance of natural language processing (NLP) algorithms in assessing RIC using head computed tomography (CT) reports written in Japanese. MATERIALS AND METHODS: 3728 Japanese head CT reports performed at Osaka University Hospital in 2020 were included. RIC (category 0: no findings, category 1: minor findings, category 2: routine follow-up, category 3: careful follow-up, and category 4: examination or therapy) was established based not only on patient severity but also on the novelty of the information. The manual assessment of RIC for the reports was performed under the consensus of two out of four neuroradiologists. The performance of four NLP models for classifying RIC was compared using fivefold cross-validation: logistic regression, bidirectional long-short-term memory (BiLSTM), general bidirectional encoder representations of transformers (general BERT), and domain-specific BERT (BERT for medical domain). RESULTS: The proportion of each RIC in the whole data set was 15.0%, 26.7%, 44.2%, 7.7%, and 6.4%, respectively. Domain-specific BERT showed the highest accuracy (0.8434 ± 0.0063) in assessing RIC and significantly higher AUC in categories 1 (0.9813 ± 0.0011), 2 (0.9492 ± 0.0045), 3 (0.9637 ± 0.0050), and 4 (0.9548 ± 0.0074) than the other models (p < .05). Analysis using layer-integrated gradients showed that the domain-specific BERT model could detect important words, such as disease names in reports. CONCLUSIONS: Domain-specific BERT has superiority over the other models in assessing our newly proposed criteria called RIC of head CT radiology reports. The accumulation of similar and further studies of has a potential to contribute to medical safety by preventing missed important findings by clinicians.

Vision-Language Model for Generating Textual Descriptions From Clinical Images: Model Development and Validation Study.

BACKGROUND: The automatic generation of radiology reports, which seeks to create a free-text description from a clinical radiograph, is emerging as a pivotal intersection between clinical medicine and artificial intelligence. Leveraging natural language processing technologies can accelerate report creation, enhancing health care quality and standardization. However, most existing studies have not yet fully tapped into the combined potential of advanced language and vision models. OBJECTIVE: The purpose of this study was to explore the integration of pretrained vision-language models into radiology report generation. This would enable the vision-language model to automatically convert clinical images into high-quality textual reports. METHODS: In our research, we introduced a radiology report generation model named ClinicalBLIP, building upon the foundational InstructBLIP model and refining it using clinical image-to-text data sets. A multistage fine-tuning approach via low-rank adaptation was proposed to deepen the semantic comprehension of the visual encoder and the large language model for clinical imagery. Furthermore, prior knowledge was integrated through prompt learning to enhance the precision of the reports generated. Experiments were conducted on both the IU X-RAY and MIMIC-CXR data sets, with ClinicalBLIP compared to several leading methods. RESULTS: Experimental results revealed that ClinicalBLIP obtained superior scores of 0.570/0.365 and 0.534/0.313 on the IU X-RAY/MIMIC-CXR test sets for the Metric for Evaluation of Translation with Explicit Ordering (METEOR) and the Recall-Oriented Understudy for Gisting Evaluation (ROUGE) evaluations, respectively. This performance notably surpasses that of existing state-of-the-art methods. Further evaluations confirmed the effectiveness of the multistage fine-tuning and the integration of prior information, leading to substantial improvements. CONCLUSIONS: The proposed ClinicalBLIP model demonstrated robustness and effectiveness in enhancing clinical radiology report generation, suggesting significant promise for real-world clinical applications.

Personalized Impression Generation for PET Reports Using Large Language Models.

Large language models (LLMs) have shown promise in accelerating radiology reporting by summarizing clinical findings into impressions. However, automatic impression generation for whole-body PET reports presents unique challenges and has received little attention. Our study aimed to evaluate whether LLMs can create clinically useful impressions for PET reporting. To this end, we fine-tuned twelve open-source language models on a corpus of 37,370 retrospective PET reports collected from our institution. All models were trained using the teacher-forcing algorithm, with the report findings and patient information as input and the original clinical impressions as reference. An extra input token encoded the reading physician's identity, allowing models to learn physician-specific reporting styles. To compare the performances of different models, we computed various automatic evaluation metrics and benchmarked them against physician preferences, ultimately selecting PEGASUS as the top LLM. To evaluate its clinical utility, three nuclear medicine physicians assessed the PEGASUS-generated impressions and original clinical impressions across 6 quality dimensions (3-point scales) and an overall utility score (5-point scale). Each physician reviewed 12 of their own reports and 12 reports from other physicians. When physicians assessed LLM impressions generated in their own style, 89% were considered clinically acceptable, with a mean utility score of 4.08/5. On average, physicians rated these personalized impressions as comparable in overall utility to the impressions dictated by other physicians (4.03, P = 0.41). In summary, our study demonstrated that personalized impressions generated by PEGASUS were clinically useful in most cases, highlighting its potential to expedite PET reporting by automatically drafting impressions.

Classification of Diagnostic Certainty in Radiology Reports with Deep Learning.

A radiology report is prepared for communicating clinical information about observed abnormal structures and clinically important findings with referring clinicians. However, such observations and findings are often accompanied by ambiguous expressions, which can prevent clinicians from accurately interpreting the content of reports. To systematically assess the degree of diagnostic certainty for each observation and finding in a report, we defined an ordinal scale comprising five classes: definite, likely, may represent, unlikely, and denial. Furthermore, we applied a deep learning classification model to determine its applicability to in-house radiology reports. We trained and evaluated the model using 540 in-house chest computed tomography reports. The deep learning model achieved a micro F1-score of 97.61%, which indicated that our ordinal scale was suitable for measuring the diagnostic certainty of observations and findings in a report.

Advanced Sampling Technique in Radiology Free-Text Data for Efficiently Building Text Mining Models by Deep Learning in Vertebral Fracture.

This study aims to establish advanced sampling methods in free-text data for efficiently building semantic text mining models using deep learning, such as identifying vertebral compression fracture (VCF) in radiology reports. We enrolled a total of 27,401 radiology free-text reports of X-ray examinations of the spine. The predictive effects were compared between text mining models built using supervised long short-term memory networks, independently derived by four sampling methods: vector sum minimization, vector sum maximization, stratified, and simple random sampling, using four fixed percentages. The drawn samples were applied to the training set, and the remaining samples were used to validate each group using different sampling methods and ratios. The predictive accuracy was measured using the area under the receiver operating characteristics (AUROC) to identify VCF. At the sampling ratios of 1/10, 1/20, 1/30, and 1/40, the highest AUROC was revealed in the sampling methods of vector sum minimization as confidence intervals of 0.981 (95%CIs: 0.980-0.983)/0.963 (95%CIs: 0.961-0.965)/0.907 (95%CIs: 0.904-0.911)/0.895 (95%CIs: 0.891-0.899), respectively. The lowest AUROC was demonstrated in the vector sum maximization. This study proposes an advanced sampling method, vector sum minimization, in free-text data that can be efficiently applied to build the text mining models by smartly drawing a small amount of critical representative samples.

Preliminary assessment of automated radiology report generation with generative pre-trained transformers: comparing results to radiologist-generated reports.

PURPOSE: In this preliminary study, we aimed to evaluate the potential of the generative pre-trained transformer (GPT) series for generating radiology reports from concise imaging findings and compare its performance with radiologist-generated reports. METHODS: This retrospective study involved 28 patients who underwent computed tomography (CT) scans and had a diagnosed disease with typical imaging findings. Radiology reports were generated using GPT-2, GPT-3.5, and GPT-4 based on the patient's age, gender, disease site, and imaging findings. We calculated the top-1, top-5 accuracy, and mean average precision (MAP) of differential diagnoses for GPT-2, GPT-3.5, GPT-4, and radiologists. Two board-certified radiologists evaluated the grammar and readability, image findings, impression, differential diagnosis, and overall quality of all reports using a 4-point scale. RESULTS: Top-1 and Top-5 accuracies for the different diagnoses were highest for radiologists, followed by GPT-4, GPT-3.5, and GPT-2, in that order (Top-1: 1.00, 0.54, 0.54, and 0.21, respectively; Top-5: 1.00, 0.96, 0.89, and 0.54, respectively). There were no significant differences in qualitative scores about grammar and readability, image findings, and overall quality between radiologists and GPT-3.5 or GPT-4 (p > 0.05). However, qualitative scores of the GPT series in impression and differential diagnosis scores were significantly lower than those of radiologists (p < 0.05). CONCLUSIONS: Our preliminary study suggests that GPT-3.5 and GPT-4 have the possibility to generate radiology reports with high readability and reasonable image findings from very short keywords; however, concerns persist regarding the accuracy of impressions and differential diagnoses, thereby requiring verification by radiologists.

Intensive vision-guided network for radiology report generation.

Objective.Automatic radiology report generation is booming due to its huge application potential for the healthcare industry. However, existing computer vision and natural language processing approaches to tackle this problem are limited in two aspects. First, when extracting image features, most of them neglect multi-view reasoning in vision and model single-view structure of medical images, such as space-view or channel-view. However, clinicians rely on multi-view imaging information for comprehensive judgment in daily clinical diagnosis. Second, when generating reports, they overlook context reasoning with multi-modal information and focus on pure textual optimization utilizing retrieval-based methods. We aim to address these two issues by proposing a model that better simulates clinicians perspectives and generates more accurate reports.Approach.Given the above limitation in feature extraction, we propose a globally-intensive attention (GIA) module in the medical image encoder to simulate and integrate multi-view vision perception. GIA aims to learn three types of vision perception: depth view, space view, and pixel view. On the other hand, to address the above problem in report generation, we explore how to involve multi-modal signals to generate precisely matched reports, i.e. how to integrate previously predicted words with region-aware visual content in next word prediction. Specifically, we design a visual knowledge-guided decoder (VKGD), which can adaptively consider how much the model needs to rely on visual information and previously predicted text to assist next word prediction. Hence, our final intensive vision-guided network framework includes a GIA-guided visual encoder and the VKGD.Main results.Experiments on two commonly-used datasets IU X-RAY and MIMIC-CXR demonstrate the superior ability of our method compared with other state-of-the-art approaches.Significance.Our model explores the potential of simulating clinicians perspectives and automatically generates more accurate reports, which promotes the exploration of medical automation and intelligence.

Management of radiology services during the 2022 FIFA football (soccer) World Cup.

Diagnostic imaging is the predominant medical service sought for the assessment and staging of musculoskeletal injuries in professional sports events. During the 2022 FIFA Football (soccer) World Cup, a centralized medical care infrastructure was established. This article provides a comprehensive account of the radiological services implemented during this event, encompassing the deployment of equipment and human resources, the structuring of workflows to uphold athlete confidentiality, and initiatives aimed at enhancing communication. Communication channels were refined through radiology consultations held with national teams' health care providers and the adoption of audiovisual reports available in multiple languages, which could be accessed remotely by team physicians. Our established framework can be replicated in international professional football events for seamless integration and efficacy.

Extracting Clinical Information From Japanese Radiology Reports Using a 2-Stage Deep Learning Approach: Algorithm Development and Validation.

Background: Radiology reports are usually written in a free-text format, which makes it challenging to reuse the reports. Objective: For secondary use, we developed a 2-stage deep learning system for extracting clinical information and converting it into a structured format. Methods: Our system mainly consists of 2 deep learning modules: entity extraction and relation extraction. For each module, state-of-the-art deep learning models were applied. We trained and evaluated the models using 1040 in-house Japanese computed tomography (CT) reports annotated by medical experts. We also evaluated the performance of the entire pipeline of our system. In addition, the ratio of annotated entities in the reports was measured to validate the coverage of the clinical information with our information model. Results: The microaveraged F1-scores of our best-performing model for entity extraction and relation extraction were 96.1% and 97.4%, respectively. The microaveraged F1-score of the 2-stage system, which is a measure of the performance of the entire pipeline of our system, was 91.9%. Our system showed encouraging results for the conversion of free-text radiology reports into a structured format. The coverage of clinical information in the reports was 96.2% (6595/6853). Conclusions: Our 2-stage deep system can extract clinical information from chest and abdomen CT reports accurately and comprehensively.

Artificial Intelligence to Improve Patient Understanding of Radiology Reports.

Diagnostic imaging reports are generally written with a target audience of other providers. As a result, the reports are written with medical jargon and technical detail to ensure accurate communication. With implementation of the 21st Century Cures Act, patients have greater and quicker access to their imaging reports, but these reports are still written above the comprehension level of the average patient. Consequently, many patients have requested reports to be conveyed in language accessible to them. Numerous studies have shown that improving patient understanding of their condition results in better outcomes, so driving comprehension of imaging reports is essential. Summary statements, second reports, and the inclusion of the radiologist's phone number have been proposed, but these solutions have implications for radiologist workflow. Artificial intelligence (AI) has the potential to simplify imaging reports without significant disruptions. Many AI technologies have been applied to radiology reports in the past for various clinical and research purposes, but patient focused solutions have largely been ignored. New natural language processing technologies and large language models (LLMs) have the potential to improve patient understanding of their imaging reports. However, LLMs are a nascent technology and significant research is required before LLM-driven report simplification is used in patient care.