Background The latest large language models (LLMs) solve unseen problems via user-defined text prompts without the need for retraining, offering potentially more efficient information extraction from free-text medical records than manual annotation. Purpose To compare the performance of the LLMs ChatGPT and GPT-4 in data mining and labeling oncologic phenotypes from free-text CT reports on lung cancer by using user-defined prompts. Materials and Methods This retrospective study included patients who underwent lung cancer follow-up CT between September 2021 and March 2023. A subset of 25 reports was reserved for prompt engineering to instruct the LLMs in extracting lesion diameters, labeling metastatic disease, and assessing oncologic progression. This output was fed into a rule-based natural language processing pipeline to match ground truth annotations from four radiologists and derive performance metrics. The oncologic reasoning of LLMs was rated on a five-point Likert scale for factual correctness and accuracy. The occurrence of confabulations was recorded. Statistical analyses included Wilcoxon signed rank and McNemar tests. Results On 424 CT reports from 424 patients (mean age, 65 years ± 11 [SD]; 265 male), GPT-4 outperformed ChatGPT in extracting lesion parameters (98.6% vs 84.0%, P < .001), resulting in 96% correctly mined reports (vs 67% for ChatGPT, P < .001). GPT-4 achieved higher accuracy in identification of metastatic disease (98.1% [95% CI: 97.7, 98.5] vs 90.3% [95% CI: 89.4, 91.0]) and higher performance in generating correct labels for oncologic progression (F1 score, 0.96 [95% CI: 0.94, 0.98] vs 0.91 [95% CI: 0.89, 0.94]) (both P < .001). In oncologic reasoning, GPT-4 had higher Likert scale scores for factual correctness (4.3 vs 3.9) and accuracy (4.4 vs 3.3), with a lower rate of confabulation (1.7% vs 13.7%) than ChatGPT (all P < .001). Conclusion When using user-defined prompts, GPT-4 outperformed ChatGPT in extracting oncologic phenotypes from free-text CT reports on lung cancer and demonstrated better oncologic reasoning with fewer confabulations. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Hafezi-Nejad and Trivedi in this issue.

“Just Accepted” papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review ...

Automated image analysis plays an increasing role in radiology in detecting and quantifying image features outside of the perception of human eyes. Common AI-based approaches address a single medical problem, although patients often present with multiple interacting, frequently subclinical medical conditions. A holistic imaging diagnostics tool based on artificial intelligence (AI) has the potential of providing an overview of multi-system comorbidities within a single workflow. An interdisciplinary, multicentric team of medical experts and computer scientists designed a pipeline, comprising AI-based tools for the automated detection, quantification and characterization of the most common pulmonary, metabolic, cardiovascular and musculoskeletal comorbidities in chest computed tomography (CT). To provide a comprehensive evaluation of each patient, a multidimensional workflow was established with algorithms operating synchronously on a decentralized Joined Imaging Platform (JIP). The results of each patient are transferred to a dedicated database and summarized as a structured report with reference to available reference values and annotated sample images of detected pathologies. Hence, this tool allows for the comprehensive, large-scale analysis of imaging-biomarkers of comorbidities in chest CT, first in science and then in clinical routine. Moreover, this tool accommodates the quantitative analysis and classification of each pathology, providing integral diagnostic and prognostic value, and subsequently leading to improved preventive patient care and further possibilities for future studies.

Low-dose four-dimensional CT was feasible for diagnosis, localization, phenotyping, and computational quantification of dynamic collapsibility of the trachea during shallow breathing; concordance with bronchoscopy was moderate and with paired CT was low.

Objectives: To evaluate the non-inferiority of non-contrast CT compared to contrast-enhanced CT with both intravenous and rectal contrast application for the diagnosis of acute colonic diverticulitis. Methods: Five readers retrospectively evaluated the non-contrast and contrast-enhanced series of CTs of 205 consecutive patients with clinical suspicion of acute diverticulitis. Two randomized reading sessions, both containing all 205 cases as either contrast-enhanced or non-contrast (1:1) series, were performed with ≥8 weeks washout between them. The non-inferiority margin was set to 0.1. Results: The pooled prevalence (all readers) of diverticulitis was similar for non-contrast CT (63.9%, range: 60.5–65.0%) and contrast-enhanced CT (64.4%, 61.5–67.8%). Non-contrast CT was non-inferior for the diagnosis of diverticulitis (accuracy 0.90 [95% confidence interval: 0.89, 0.92]) compared to contrast-enhanced CT (0.92 [0.90, 0.94]; the difference in accuracy: −0.01 [−0.04, 0.01]) (normal deviate test: p-value_one-sided = 5.20 × 10⁻⁶). Sensitivities for perforation and abscess were slightly but significantly lower for the non-contrast CT than for the contrast-enhanced CT (differences: −0.15 [−0.20, −0.05], −0.17 [−0.27, −0.07]), while no differences in accuracies and specificities were observed. Conclusions: Non-contrast CT is non-inferior to contrast-enhanced CT (intravenous and rectal contrast) for the diagnosis of acute colonic diverticulitis. Contrast-enhanced CT is associated with significantly higher sensitivities for the presence of an abscess or perforation.