Radiology Reports Research Articles

Preliminary assessment of TNM classification performance for pancreatic cancer in Japanese radiology reports using GPT-4.

A large-scale language model is expected to have been trained with a large volume of data including cancer treatment protocols. The current study aimed to investigate the use of generative pretrained transformer 4 (GPT-4) for identifying the TNM classification of pancreatic cancers from existing radiology reports written in Japanese. We screened 100 consecutive radiology reports on computed tomography scan for pancreatic cancer from April 2020 to June 2022. GPT-4 was requested to classify the TNM from the radiology reports based on the General Rules for the Study of Pancreatic Cancer 7th Edition. The accuracy and kappa coefficient of the TNM classifications by GPT-4 was evaluated with the classifications by two experienced abdominal radiologists as gold standard. The accuracy values of the T, N, and M factors were 0.73, 0.91, and 0.93, respectively. The kappa coefficients were 0.45 for T, 0.79 for N, and 0.83 for M. Although GPT is familiar with the TNM classification for pancreatic cancer, its performance in classifying actual cases in this experiment may not be adequate.

DKA-RG: Disease-Knowledge-Enhanced Fine-Grained Image–Text Alignment for Automatic Radiology Report Generation

Automatic radiology report generation is a task that combines artificial intelligence and medical information processing, and it fully relies on computer vision and natural language processing techniques. Nowadays, automatic radiology report generation is still a very challenging task because it requires semantically adequate alignment of data from two modalities: radiology images and text. Existing approaches tend to focus on coarse-grained alignment at the global level and do not take into account the disease characteristics of radiology images at fine-grained semantics, which results in the generated reports potentially omitting key disease diagnostic descriptions. In this work, we propose a new approach, disease-knowledge-enhanced fine-grained image–text alignment for automatic radiology report generation (DKA-RG). The method combines global and disease-level alignment, thus facilitating the extraction of fine-grained disease features by the model. Our approach also introduces a knowledge graph to inject medical domain expertise into the model. Our proposed DKA-RG consists of two training steps: the image–report alignment stage and the image-to-report generation stage. In the alignment stage, we use global contrastive learning to align images and texts from a high level and also augment disease contrastive learning with medical knowledge to enhance the disease detection capability. In the report generation stage, the report text generated from the images is more accurate in describing the disease information thanks to sufficient alignment. Through extensive quantitative and qualitative experiments on two widely used datasets, we validate the effectiveness of our DKA-RG on the task of radiology report generation. Our DKA-RG achieves superior performance on multiple types of metrics (natural language generation and clinical efficacy metrics) compared to existing methods, demonstrating that the method can improve the reliability and accuracy of automatic radiology report generation systems.

From Revisions to Insights: Converting Radiology Report Revisions into Actionable Educational Feedback Using Generative AI Models.

Expert feedback on trainees' preliminary reports is crucial for radiologic training, but real-time feedback can be challenging due to non-contemporaneous, remote reading and increasing imaging volumes. Trainee report revisions contain valuable educational feedback, but synthesizing data from raw revisions is challenging. Generative AI models can potentially analyze these revisions and provide structured, actionable feedback. This study used the OpenAI GPT-4 Turbo API to analyze paired synthesized and open-source analogs of preliminary and finalized reports, identify discrepancies, categorize their severity and type, and suggest review topics. Expert radiologists reviewed the output by grading discrepancies, evaluating the severity and category accuracy, and suggested review topic relevance. The reproducibility of discrepancy detection and maximal discrepancy severity was also examined. The model exhibited high sensitivity, detecting significantly more discrepancies than radiologists (W = 19.0, p < 0.001) with a strong positive correlation (r = 0.778, p < 0.001). Interrater reliability for severity and type were fair (Fleiss' kappa = 0.346 and 0.340, respectively; weighted kappa = 0.622 for severity). The LLM achieved a weighted F1 score of 0.66 for severity and 0.64 for type. Generated teaching points were considered relevant in ~ 85% of cases, and relevance correlated with the maximal discrepancy severity (Spearman ρ = 0.76, p < 0.001). The reproducibility was moderate to good (ICC (2,1) = 0.690) for the number of discrepancies and substantial for maximal discrepancy severity (Fleiss' kappa = 0.718; weighted kappa = 0.94). Generative AI models can effectively identify discrepancies in report revisions and generate relevant educational feedback, offering promise for enhancing radiology training.

Rethinking masked image modelling for medical image representation

Masked Image Modelling (MIM), a form of self-supervised learning, has garnered significant success in computer vision by improving image representations using unannotated data. Traditional MIMs typically employ a strategy of random sampling across the image. However, this random masking technique may not be ideally suited for medical imaging, which possesses distinct characteristics divergent from natural images. In medical imaging, particularly in pathology, disease-related features are often exceedingly sparse and localized, while the remaining regions appear normal and undifferentiated. Additionally, medical images frequently accompany reports, directly pinpointing pathological changes’ location. Inspired by this, we propose Masked medical Image Modelling (MedIM), a novel approach, to our knowledge, the first research that employs radiological reports to guide the masking and restore the informative areas of images, encouraging the network to explore the stronger semantic representations from medical images. We introduce two mutual comprehensive masking strategies, knowledge-driven masking (KDM), and sentence-driven masking (SDM). KDM uses Medical Subject Headings (MeSH) words unique to radiology reports to identify symptom clues mapped to MeSH words (e.g., cardiac, edema, vascular, pulmonary) and guide the mask generation. Recognizing that radiological reports often comprise several sentences detailing varied findings, SDM integrates sentence-level information to identify key regions for masking. MedIM reconstructs images informed by this masking from the KDM and SDM modules, promoting a comprehensive and enriched medical image representation. Our extensive experiments on seven downstream tasks covering multi-label/class image classification, pneumothorax segmentation, and medical image–report analysis, demonstrate that MedIM with report-guided masking achieves competitive performance. Our method substantially outperforms ImageNet pre-training, MIM-based pre-training, and medical image–report pre-training counterparts. Codes are available at https://github.com/YtongXie/MedIM.

Decoding Radiology Reports: Artificial Intelligence-Large Language Models Can Improve the Readability of Hand and Wrist Orthopedic Radiology Reports.

The purpose of this study was to assess the effectiveness of an Artificial Intelligence-Large Language Model (AI-LLM) at improving the readability of hand and wrist radiology reports. The radiology reports of 100 hand and/or wrist radiographs, 100 hand and/or wrist computed tomography (CT) scans, and 100 hand and/or wrist magnetic resonance imaging (MRI) scans were extracted. The following prompt command was inserted into the AI-LLM: "Explain this radiology report to a patient in layman's terms in the second person: [Report Text]." The report length, Flesch reading ease score (FRES), and Flesch-Kincaid reading level (FKRL) were calculated for the original radiology report and the AI-LLM-generated report. The accuracy of the AI-LLM report was assessed via a 5-point Likert scale. Any "hallucination" produced by the AI-LLM-generated report was recorded. There was a statistically significant improvement in mean FRES scores and FKRL scores in the AI-LLM-generated radiograph report, CT report, and MRI report. For all AI-LLM-generated reports, the mean reading level improved to below an eighth-grade reading level. The mean Likert score for the AI-LLM-generated radiograph report, CT report, and MRI report was 4.1 ± 0.6, 3.9 ± 0.6, and 3.9 ± 0.7, respectively. The hallucination rate in the AI-LLM-generated radiograph report, CT report, and MRI report was 3%, 6%, and 6%, respectively. This study demonstrates that AI-LLM effectively improves the readability of hand and wrist radiology reports, underscoring the potential application of AI-LLM as a promising and innovative patient-centric strategy to improve patient comprehension of their imaging reports.Level of Evidence: IV.

Enhancing the vision–language foundation model with key semantic knowledge-emphasized report refinement

Recently, vision–language representation learning has made remarkable advancements in building up medical foundation models, holding immense potential for transforming the landscape of clinical research and medical care. The underlying hypothesis is that the rich knowledge embedded in radiology reports can effectively assist and guide the learning process, reducing the need for additional labels. However, these reports tend to be complex and sometimes even consist of redundant descriptions that make the representation learning too challenging to capture the key semantic information. This paper develops a novel iterative vision–language representation learning framework by proposing a key semantic knowledge-emphasized report refinement method. Particularly, raw radiology reports are refined to highlight the key information according to a constructed clinical dictionary and two model-optimized knowledge-enhancement metrics. The iterative framework is designed to progressively learn, starting from gaining a general understanding of the patient’s condition based on raw reports and gradually refines and extracts critical information essential to the fine-grained analysis tasks. The effectiveness of the proposed framework is validated on various downstream medical image analysis tasks, including disease classification, region-of-interest segmentation, and phrase grounding. Our framework surpasses seven state-of-the-art methods in both fine-tuning and zero-shot settings, demonstrating its encouraging potential for different clinical applications.

Scar Decompression in Managing Breast Cancer-Related Lymphedema: Is it Needed?

Mastectomy, axillary lymph node dissection, and irradiation for breast cancer commonly result in perivascular and axillary scarring. This scarring is thought to cause functional venous stenosis that leads to downstream venous hypertension in the affected extremity. Standard surgical practice is to decompress perivascular scarring at the time of physiologic lymphedema surgery in patients with breast cancer-related lymphedema (BCRL). However, it is unknown whether this scar release influences surgical outcomes. The purpose of this study was to evaluate the prevalence of functional venous stenosis in patients with BCRL and determine whether scar decompression is a necessary step in physiologic lymphedema surgery. The authors conducted a retrospective review of 64 patients with unilateral BCRL that presented to our lymphedema center between January 2020 and October 2022. Radiologist reports of venous duplex ultrasound for the bilateral upper extremities identified any disturbances in venous flow or indications of venous stenosis. Of the 64 patients with BCRL, 78% (n = 50) had prior axillary lymph node dissection. Forty-seven (73%) patients completed ultrasound imaging, of which, one patient (2%) had venous stenosis in the affected lymphedematous extremity identified on duplex ultrasound that may have suggested functional scarring. Vascularized lymph node transfer (VLNT) without scar decompression was performed in six patients (9%). Average preoperative Lymphedema Life Impact Scale and Lymphedema Index scores were 35 and 19 units, with a mean decrease of 23 (67%) and 6 (30%) units postoperatively. Most patients with BCRL did not have identifiable functional venous stenosis on duplex ultrasound, apart from one patient with suspected postthrombotic changes. All six patients that received VLNT without scar decompression had a successful outcome with decreased measures of lymphedema postoperatively. Scar decompression may therefore be unnecessary in physiologic lymphedema surgery, reducing operative times and avoiding risk of injury to neurovascular structures of the axilla.

Automatic ICD-10 coding : Natural language processing for German MRI reports

The medical coding of radiology reports is essential for a good quality of care and correct billing, but at the same time acomplex and error-prone task. To assess the performance of natural language processing (NLP) for ICD-10 coding of German radiology reports using fine tuning of suitable language models. This retrospective study included all magnetic resonance imaging (MRI) radiology reports acquired at our institution between 2010 and 2020. The codes on discharge ICD-10 were matched to the corresponding reports to construct adataset for multiclass classification. Fine tuning of GermanBERT and flanT5 was carried out on the total dataset (dstotal) containing 1035 different ICD-10codes and 2 reduced subsets containing the 100 (ds100) and 50 (ds50) most frequent codes. The performance of the model was assessed using top‑k accuracy for k = 1, 3 and 5. In an ablation study both models were trained on the accompanying metadata and the radiology report alone. The total dataset consisted of 100,672 radiology reports, the reduced subsets ds100 of 68,103 and ds50 of 52,293 reports. The performance of the model increased when several of the best predictions of the model were taken into consideration, when the number of target classes was reduced and the metadata were combined with the report. The flanT5 outperformed GermanBERT across all datasets and metrics and was is suited as a medical coding assistant, achieving atop 3 accuracy of nearly 70% in the real-world dataset dstotal. Finely tuned language models can reliably predict ICD-10codes of German magnetic resonance imaging (MRI) radiology reports across various settings. As acoding assistant flanT5 can guide medical coders to make informed decisions and potentially reduce the workload.

A generalist vision-language foundation model for diverse biomedical tasks.

Traditional biomedical artificial intelligence (AI) models, designed for specific tasks or modalities, often exhibit limited flexibility in real-world deployment and struggle to utilize holistic information. Generalist AI holds the potential to address these limitations due to its versatility in interpreting different data types and generating tailored outputs for diverse needs. However, existing biomedical generalist AI solutions are typically heavyweight and closed source to researchers, practitioners and patients. Here, we describe BiomedGPT, the first open-source and lightweight vision-language foundation model, designed as a generalist capable of performing various biomedical tasks. BiomedGPT achieved state-of-the-art results in 16 out of 25 experiments while maintaining a computing-friendly model scale. We also conducted human evaluations to assess the capabilities of BiomedGPT in radiology visual question answering, report generation and summarization. BiomedGPT exhibits robust prediction ability with a low error rate of 3.8% in question answering, satisfactory performance with an error rate of 8.3% in writing complex radiology reports, and competitive summarization ability with a nearly equivalent preference score to human experts. Our method demonstrates that effective training with diverse data can lead to more practical biomedical AI for improving diagnosis and workflow efficiency.

IQAGPT: computed tomography image quality assessment with vision-language and ChatGPT models

Large language models (LLMs), such as ChatGPT, have demonstrated impressive capabilities in various tasks and attracted increasing interest as a natural language interface across many domains. Recently, large vision-language models (VLMs) that learn rich vision–language correlation from image–text pairs, like BLIP-2 and GPT-4, have been intensively investigated. However, despite these developments, the application of LLMs and VLMs in image quality assessment (IQA), particularly in medical imaging, remains unexplored. This is valuable for objective performance evaluation and potential supplement or even replacement of radiologists’ opinions. To this end, this study introduces IQAGPT, an innovative computed tomography (CT) IQA system that integrates image-quality captioning VLM with ChatGPT to generate quality scores and textual reports. First, a CT-IQA dataset comprising 1,000 CT slices with diverse quality levels is professionally annotated and compiled for training and evaluation. To better leverage the capabilities of LLMs, the annotated quality scores are converted into semantically rich text descriptions using a prompt template. Second, the image-quality captioning VLM is fine-tuned on the CT-IQA dataset to generate quality descriptions. The captioning model fuses image and text features through cross-modal attention. Third, based on the quality descriptions, users verbally request ChatGPT to rate image-quality scores or produce radiological quality reports. Results demonstrate the feasibility of assessing image quality using LLMs. The proposed IQAGPT outperformed GPT-4 and CLIP-IQA, as well as multitask classification and regression models that solely rely on images.

Deep learning reconstruction for optimized bone assessment in zero echo time MR imaging of the knee

PurposeTo evaluate the impact of deep learning-based reconstruction (DLRecon) on bone assessment in zero echo-time (ZTE) MRI of the knee at 1.5 Tesla. MethodsThis retrospective study included 48 consecutive exams of 46 patients (23 females) who underwent clinically indicated knee MRI at 1.5 Tesla. Standard imaging protocol comprised a sagittal prescribed, isotropic ZTE sequence. ZTE image reconstruction was performed with a standard-of-care (non-DL) and prototype DLRecon method. Exams were divided into subsets with and without osseous pathology based on the radiology report. Using a 4-point scale, two blinded readers qualitatively graded features of bone depiction including artifacts and conspicuity of pathology including diagnostic certainty in the respective subsets. Quantitatively, one reader measured signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of bone. Comparative analyses were conducted to assess the differences between the reconstruction methods. In addition, interreader agreement was calculated for the qualitative gradings. ResultsDLRecon significantly improved gradings for bone depiction relative to non-DL reconstruction (all, p < 0.05), while there was no significant difference with regards to artifacts (both, median score of 0; p = 0.058). In the subset with pathologies, conspicuity of pathology and diagnostic confidence were also scored significantly higher in DLRecon compared to non-DL (median 3 vs 2; p ≤ 0.03). Interreader agreement ranged from moderate to almost-perfect (κ = 0.54–0.88). Quantitatively, DLRecon demonstrated significantly enhanced CNR and SNR of bone compared to non-DL (p < 0.001). ConclusionZTE MRI with DLRecon improved bone depiction in the knee, compared to non-DL. Additionally, DLRecon increased conspicuity of osseous findings together with diagnostic certainty.

Performance of AI for preoperative CT assessment of lung metastases: Retrospective analysis of 167 patients

ObjectivesTo evaluate the performance of artificial intelligence (AI) in the preoperative detection of lung metastases on CT. Materials and methodsPatients who underwent lung metastasectomy in our institution between 2016 and 2020 were enrolled, their preoperative CT reports having been performed before an AI solution (Veye Lung Nodules, version 3.9.2, Aidence) became available as a second reader in our department.All CT scans were retrospectively processed by AI. The sensitivities of unassisted radiologists (original CT radiology reports), AI reports alone and both combined were compared. Ground truth was established by a consensus reading of two radiologists, who analyzed whether the nodules mentioned in the pathology report were retrospectively visible on CT. Multivariate analysis was performed to identify nodule characteristics associated with detectability. ResultsA total of 167 patients (men: 62.9 %; median age, 59 years [47–68]) with 475 resected nodules were included. AI detected an average of 4 nodules (0–17) per CT, of which 97 % were true nodules. The combination of radiologist plus AI (92.4 %) had significantly higher sensitivity than unassisted radiologists (80.4 %) (p < 0.001). In 27/57 (47.4 %) patients who had multiple preoperative CT scans, AI detected lung nodules earlier than the radiologist. Vascular contact was associated with non-detection by radiologists (OR:0.32[0.19, 0.54], p < 0.001), whilst the presence of cavitation (OR:0.26[0.13, 0.54], p < 0.001) or pleural contact (OR:0.10[0.04, 0.22], p < 0.001) was associated with non-detection by AI. ConclusionAI significantly increases the sensitivity of preoperative detection of lung metastases and enables earlier detection, with a significant potential benefit for patient management.

Using AI to Identify Unremarkable Chest Radiographs for Automatic Reporting.

Background Radiology practices have a high volume of unremarkable chest radiographs and artificial intelligence (AI) could possibly improve workflow by providing an automatic report. Purpose To estimate the proportion of unremarkable chest radiographs, where AI can correctly exclude pathology (ie, specificity) without increasing diagnostic errors. Materials and Methods In this retrospective study, consecutive chest radiographs in unique adult patients (≥18 years of age) were obtained January 1-12, 2020, at four Danish hospitals. Exclusion criteria included insufficient radiology reports or AI output error. Two thoracic radiologists, who were blinded to AI output, labeled chest radiographs as "remarkable" or "unremarkable" based on predefined unremarkable findings (reference standard). Radiology reports were classified similarly. A commercial AI tool was adapted to output a chest radiograph "remarkableness" probability, which was used to calculate specificity at different AI sensitivities. Chest radiographs with missed findings by AI and/or the radiology report were graded by one thoracic radiologist as critical, clinically significant, or clinically insignificant. Paired proportions were compared using the McNemar test. Results A total of 1961 patients were included (median age, 72 years [IQR, 58-81 years]; 993 female), with one chest radiograph per patient. The reference standard labeled 1231 of 1961 chest radiographs (62.8%) as remarkable and 730 of 1961 (37.2%) as unremarkable. At 99.9%, 99.0%, and 98.0% sensitivity, the AI had a specificity of 24.5% (179 of 730 radiographs [95% CI: 21, 28]), 47.1% (344 of 730 radiographs [95% CI: 43, 51]), and 52.7% (385 of 730 radiographs [95% CI: 49, 56]), respectively. With the AI fixed to have a similar sensitivity as radiology reports (87.2%), the missed findings of AI and reports had 2.2% (27 of 1231 radiographs) and 1.1% (14 of 1231 radiographs) classified as critical (P = .01), 4.1% (51 of 1231 radiographs) and 3.6% (44 of 1231 radiographs) classified as clinically significant (P = .46), and 6.5% (80 of 1231) and 8.1% (100 of 1231) classified as clinically insignificant (P = .11), respectively. At sensitivities greater than or equal to 95.4%, the AI tool exhibited less than or equal to 1.1% critical misses. Conclusion A commercial AI tool used off-label could correctly exclude pathology in 24.5%-52.7% of all unremarkable chest radiographs at greater than or equal to 98% sensitivity. The AI had equal or lower rates of critical misses than radiology reports at sensitivities greater than or equal to 95.4%. These results should be confirmed in a prospective study. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Yoon and Hwang in this issue.

Knowledge-Augmented Deep Learning for Segmenting and Detecting Cerebral Aneurysms With CT Angiography: A Multicenter Study.

Background Deep learning (DL) could improve the labor-intensive, challenging processes of diagnosing cerebral aneurysms but requires large multicenter data sets. Purpose To construct a DL model using a multicenter data set for accurate cerebral aneurysm segmentation and detection on CT angiography (CTA) images and to compare its performance with radiology reports. Materials and Methods Consecutive head or head and neck CTA images of suspected unruptured cerebral aneurysms were gathered retrospectively from eight hospitals between February 2018 and October 2021 for model development. An external test set with reference standard digital subtraction angiography (DSA) scans was obtained retrospectively from one of the eight hospitals between February 2022 and February 2023. Radiologists (reference standard) assessed aneurysm segmentation, while model performance was evaluated using the Dice similarity coefficient (DSC). The model's aneurysm detection performance was assessed by sensitivity and comparing areas under the receiver operating characteristic curves (AUCs) between the model and radiology reports in the DSA data set with use of the DeLong test. Results Images from 6060 patients (mean age, 56 years ± 12 [SD]; 3375 [55.7%] female) were included for model development (training: 4342; validation: 1086; and internal test set: 632). Another 118 patients (mean age, 59 years ± 14; 79 [66.9%] female) were included in an external test set to evaluate performance based on DSA. The model achieved a DSC of 0.87 for aneurysm segmentation performance in the internal test set. Using DSA, the model achieved 85.7% (108 of 126 aneurysms [95% CI: 78.1, 90.1]) sensitivity in detecting aneurysms on per-vessel analysis, with no evidence of a difference versus radiology reports (AUC, 0.93 [95% CI: 0.90, 0.95] vs 0.91 [95% CI: 0.87, 0.94]; P = .67). Model processing time from reconstruction to detection was 1.76 minutes ± 0.32 per scan. Conclusion The proposed DL model could accurately segment and detect cerebral aneurysms at CTA with no evidence of a significant difference in diagnostic performance compared with radiology reports. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Payabvash in this issue.

Enhancing Radiological Reporting in Head and Neck Cancer: Converting Free-Text CT Scan Reports to Structured Reports Using Large Language Models

Objective The aim of this study was to assess efficacy of large language models (LLMs) for converting free-text computed tomography (CT) scan reports of head and neck cancer (HNCa) patients into a structured format using a predefined template. Materials and Methods A retrospective study was conducted using 150 CT reports of HNCa patients. A comprehensive structured reporting template for HNCa CT scans was developed, and the Generative Pre-trained Transformer 4 (GPT-4) was initially used to convert 50 CT reports into a structured format using this template. The generated structured reports were then evaluated by a radiologist for instances of missing or misinterpreted information and any erroneous additional details added by GPT-4. Following this assessment, the template was refined for improved accuracy. This revised template was then used for conversion of 100 other HNCa CT reports into structured format using GPT-4. These reports were then reevaluated in the same manner. Results Initially, GPT-4 successfully converted all 50 free-text reports into structured reports. However, there were 10 places with missing information: tracheostomy tube (n = 3), noninclusion of involvement of sternocleidomastoid muscle (n = 2), extranodal tumor extension (n = 3), and contiguous involvement of the neck structures by nodal mass rather than the primary (n = 2). Few instances of nonsuspicious lung nodules were misinterpreted as metastases (n = 2). GPT-4 did not indicate any erroneous additional findings. Using the revised reporting template, GPT-4 converted all the 100 CT reports into a structured format with no repeated or additional mistakes. Conclusion LLMs can be used for structuring free-text radiology reports using plain language prompts and a simple yet comprehensive reporting template. Summary Statement Large language models can successfully and accurately convert conventional radiology reports for oncology scans into a structured format using a comprehensive predefined template and thus can enhance the utility and integration of these reports in routine clinical practice. Key Points

An open chest X-ray dataset with benchmarks for automatic radiology report generation in French

Medical report generation (MRG), which aims to automatically generate a textual description of a specific medical image (e.g., a chest X-ray), has recently received increasing research interest. Building on the success of image captioning, MRG has become achievable. However, generating language-specific radiology reports poses a challenge for data-driven models due to their reliance on paired image-report chest X-ray datasets, which are labor-intensive, time-consuming, and costly. In this paper, we introduce a chest X-ray benchmark dataset, namely CASIA-CXR, consisting of high-resolution chest radiographs accompanied by narrative reports originally written in French. To the best of our knowledge, this is the first public chest radiograph dataset with medical reports in this particular language. Importantly, we propose a simple yet effective multimodal encoder–decoder contextually-guided framework for medical report generation in French. We validated our framework through intra-language and cross-language contextual analysis, supplemented by expert evaluation performed by radiologists. The dataset is freely available at: https://www.casia-cxr.net/.

Fully Automated Deep Learning Model to Detect Clinically Significant Prostate Cancer at MRI.

Background Multiparametric MRI can help identify clinically significant prostate cancer (csPCa) (Gleason score ≥7) but is limited by reader experience and interobserver variability. In contrast, deep learning (DL) produces deterministic outputs. Purpose To develop a DL model to predict the presence of csPCa by using patient-level labels without information about tumor location and to compare its performance with that of radiologists. Materials and Methods Data from patients without known csPCa who underwent MRI from January 2017 to December 2019 at one of multiple sites of a single academic institution were retrospectively reviewed. A convolutional neural network was trained to predict csPCa from T2-weighted images, diffusion-weighted images, apparent diffusion coefficient maps, and T1-weighted contrast-enhanced images. The reference standard was pathologic diagnosis. Radiologist performance was evaluated as follows: Radiology reports were used for the internal test set, and four radiologists' PI-RADS ratings were used for the external (ProstateX) test set. The performance was compared using areas under the receiver operating characteristic curves (AUCs) and the DeLong test. Gradient-weighted class activation maps (Grad-CAMs) were used to show tumor localization. Results Among 5735 examinations in 5215 patients (mean age, 66 years ± 8 [SD]; all male), 1514 examinations (1454 patients) showed csPCa. In the internal test set (400 examinations), the AUC was 0.89 and 0.89 for the DL classifier and radiologists, respectively (P = .88). In the external test set (204 examinations), the AUC was 0.86 and 0.84 for the DL classifier and radiologists, respectively (P = .68). DL classifier plus radiologists had an AUC of 0.89 (P < .001). Grad-CAMs demonstrated activation over the csPCa lesion in 35 of 38 and 56 of 58 true-positive examinations in internal and external test sets, respectively. Conclusion The performance of a DL model was not different from that of radiologists in the detection of csPCa at MRI, and Grad-CAMs localized the tumor. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Johnson and Chandarana in this issue.

Machine learning and deep learning tools for the automated capture of cancer surveillance data.

The National Cancer Institute and the Department of Energy strategic partnership applies advanced computing and predictive machine learning and deep learning models to automate the capture of information from unstructured clinical text for inclusion in cancer registries. Applications include extraction of key data elements from pathology reports, determination of whether a pathology or radiology report is related to cancer, extraction of relevant biomarker information, and identification of recurrence. With the growing complexity of cancer diagnosis and treatment, capturing essential information with purely manual methods is increasingly difficult. These new methods for applying advanced computational capabilities to automate data extraction represent an opportunity to close critical information gaps and create a nimble, flexible platform on which new information sources, such as genomics, can be added. This will ultimately provide a deeper understanding of the drivers of cancer and outcomes in the population and increase the timeliness of reporting. These advances will enable better understanding of how real-world patients are treated and the outcomes associated with those treatments in the context of our complex medical and social environment.

Leveraging the full expertise of radiographers for improved healthcare delivery.

The role of radiographers in healthcare has evolved significantly from operating imaging equipment to being essential in patient care and diagnosis. In Australia, radiographers play a crucial role in image interpretation, identifying and communicating significant findings to enhance patient outcomes. Preliminary image evaluation (PIE) allows radiographers to interpret images and ensure significant findings are noted, particularly in urgent situations, complementing diagnoses when radiologist reports are unavailable. Despite their potential, many radiographers lack empowerment, leading to delays and adverse patient outcomes. This underutilisation stems from a lack of support and systemic barriers. Radiographers, with their extensive expertise in imaging, are vital for ensuring patient safety and care quality. Policy changes are needed to integrate PIE into standard workflows, allowing radiographers to fully utilise their skills. Recognising and leveraging their expertise will enhance patient care, foster collaboration, and ensure radiographers contribute fully to the healthcare team, ultimately improving patient safety and care quality.

Associations Between Chronic Rhinosinusitis and the Development of Non–Cystic Fibrosis Bronchiectasis

BackgroundStudies have shown an association between chronic rhinosinusitis (CRS) and non-cystic fibrosis (CF) bronchiectasis. ObjectiveWe aimed to determine if CRS increases the risk of developing non-CF bronchiectasis. MethodsA retrospective analysis was conducted utilizing electronic medical records from an academic center. Patients with CRS without bronchiectasis, with at least one chest computed tomography (CT) performed after the diagnosis of CRS, were identified between January 2006 and December 2015. Charts were reviewed until May 2022. The control group was age, sex, and race matched, and included patients without CRS, asthma, or chronic obstructive pulmonary disease (COPD) who had at least one chest CT. Bronchiectasis was identified by chest CT radiology reports. The odds of developing bronchiectasis were analyzed in patients with CRS without asthma or COPD (Cohort 1) and patients with CRS with asthma or COPD (Cohort 2). ResultsThe odds of developing bronchiectasis were significantly higher in patients with CRS (139/1594, 8.7%) compared to patients in the control group (443/7992 [5.5%], OR 1.63 [1.34-1.99]). Furthermore, the odds of developing bronchiectasis were higher in Cohort 1 (63/863 [7.3%], OR 1.34 [1.02-1.76]) and Cohort 2 (76/731 [10.4%], OR 1.98 [1.53-2.55]) versus the control group. After adjusting for confounding diseases, the association was attenuated in Cohort 1 (OR 1.22 [0.92-1.61]) but remained significant in Cohort 2 (OR 1.78 [1.37-2.31]). ConclusionsCRS is associated with the future development of non-CF bronchiectasis. Patients with CRS, especially those with asthma or COPD, have a higher likelihood of developing bronchiectasis than patients without CRS.