BACKGROUND AND PURPOSE
Artificial intelligence (AI) algorithms have been used to synthesize standard-dose images from low-dose images in brain MRI, but have less been evaluated to boost standard-dose contrast to approximate higher-dose effect. This study aims to evaluate the performance of a deep learning-based post-processing tool (AiMIFY) in enhancing contrast and improving lesion visualization on standard-dose contrast-enhanced brain MRI.
MATERIALS AND METHODS
In this retrospective, multicenter, multireader study, 86 adult patients who underwent standard-dose contrast-enhanced brain MRI were included. Pre-contrast and post-contrast three-dimensional T1-weighted images were processed using AiMIFY to generate contrast-boosting images. Three independent neuroradiologists performed blinded assessments. Quantitative metrics, including contrast-to-noise ratio (CNR), lesion-to-brain ratio (LBR), and contrast enhancement percentage (CEP), were measured. Subjective image quality (border delineation, internal morphology, and contrast enhancement) were evaluated using a Likert scale. The overall diagnostic preference was recorded. Comparisons between standard post-contrast and AiMIFY-processed images were performed using the Wilcoxon signed-rank test.
RESULTS
AiMIFY-processed images demonstrated significantly higher CNR, LBR, and CEP compared with standard-dose post-contrast images across all readers (all P < 0.001), with mean increases of 495.16%, 58.94%, and 160.55%, respectively. Subjective assessments showed significant improvements in border delineation, internal morphology, and contrast enhancement (all P < 0.05). Subgroup analysis of small lesions (< 10 mm) revealed consistently higher subjective scores for AiMIFY-processed images (all P < 0.05). AiMIFY-processed images were preferred in the majority of cases (84.9%, 29.1%, and 68.6% across readers; all P < 0.001).
CONCLUSION
Deep learning-based enhancement using AiMIFY significantly improves lesion conspicuity on standard-dose contrast-enhanced brain MRI, with notable benefits for small lesions. This approach may represent an adjuvant to standard-dose MRI to enhance diagnostic confidence while avoiding increased contrast agent exposure.
Yuantong Gao, Bin Chen, Caiyun Wen et al.· AJNR. American journal of ne...· 0 citations
Abstract Background Large language models (LLMs) show promise in automatically detecting errors in radiology reports, but their performance remains insufficiently validated in large-scale, real-world clinical datasets. Objective This study aimed to systematically evaluate the performance of LLMs in detecting and correcting errors in Chinese radiology reports derived from authentic clinical data. Methods A large-scale dataset of 4480 Chinese radiology reports with modification records containing real clinical practice-generated errors was retrospectively collected between January 2023 and June 2024 at a single institution. After exclusions, 1363 reports containing 1551 errors were included. The dataset covers various anatomical parts of the body from different imaging modalities and was randomly divided into a test set (n=1263) and an internal validation set (n=100). Additionally, 100 error-free reports were added to the internal validation set. An additional 200 English-language reports from the Medical Information Mart for Intensive Care (MIMIC-III) were used for external validation. Eight human readers and 8 widely adopted LLMs, enhanced by prompt engineering, were tasked with error detection. Overall and subgroup detection performance and reading time were evaluated. Correction suggestions from the 2 best-performing LLMs were reviewed by a senior radiologist. Results On the test set, DeepSeek-R1 achieved the highest overall detection rate at 89% (95% CI 87%-90%), significantly better than the other 7 models (P=.001-.007). On the internal validation set, DeepSeek-R1 and Claude-3.5-Sonnet achieved detection rates of 83% (100/120; 95% CI 76%-89%) and 80% (96/120; 95% CI 72%-86%), respectively. DeepSeek-R1 showed performance comparable to radiologists (83%, 95% CI 76%-89% vs 80%, 95% CI 72%-86% for junior radiologists and 78%, 95% CI 70%-85% for senior radiologists; P=.39 and P=.19, respectively) and significantly better performance than that of nonradiologists and nonphysicians (83%, 95% CI 76%-89% vs 66%, 95% CI 57%-74% and 38%, 95% CI 30%-47%; P<.001, respectively). DeepSeek-R1 showed a false-positive rate comparable to radiologists (DeepSeek-R1 vs senior radiologists and junior radiologists, 3% vs 0% and 1%; P=.25 and P=.61, respectively) and a significantly lower rate than nonradiologists and nonphysicians (3% vs 13% and 17%; P=.02 and P=.002, respectively). On the external validation set, DeepSeek-R1 and Claude-3.5-Sonnet achieved detection rates of 94% (95% CI 89%-97%) and 93% (95% CI 88%-97%), respectively. The correction accuracy of DeepSeek-R1 and Claude-3.5-Sonnet was 95% and 91%, respectively. Conclusions Enhanced LLMs, particularly DeepSeek-R1, demonstrated robust performance in error detection and correction within real-world Chinese radiology reports, supporting their clinical use for automated quality assurance and integration into workflows to improve reporting accuracy and efficiency.
Jiafeng Zhou, Yuxin Wei, Qian Cai et al.· Journal of Medical Internet...· 0 citations