Evaluating GPT-V4 (GPT-4 with Vision) on Detection of Radiologic Findings on Chest Radiographs.

Background Generating radiologic findings from chest radiographs is pivotal in medical image analysis. The emergence of OpenAI's generative pretrained transformer, GPT-4 with vision (GPT-4V), has opened new perspectives on the potential for automated image-text pair generation. However, the application of GPT-4V to real-world chest radiography is yet to be thoroughly examined. Purpose To investigate the capability of GPT-4V to generate radiologic findings from real-world chest radiographs. Materials and Methods In this retrospective study, 100 chest radiographs with free-text radiology reports were annotated by a cohort of radiologists, two attending physicians and three residents, to establish a reference standard. Of 100 chest radiographs, 50 were randomly selected from the National Institutes of Health (NIH) chest radiographic data set, and 50 were randomly selected from the Medical Imaging and Data Resource Center (MIDRC). The performance of GPT-4V at detecting imaging findings from each chest radiograph was assessed in the zero-shot setting (where it operates without prior examples) and few-shot setting (where it operates with two examples). Its outcomes were compared with the reference standard with regards to clinical conditions and their corresponding codes in the International Statistical Classification of Diseases, Tenth Revision (ICD-10), including the anatomic location (hereafter, laterality). Results In the zero-shot setting, in the task of detecting ICD-10 codes alone, GPT-4V attained an average positive predictive value (PPV) of 12.3%, average true-positive rate (TPR) of 5.8%, and average F1 score of 7.3% on the NIH data set, and an average PPV of 25.0%, average TPR of 16.8%, and average F1 score of 18.2% on the MIDRC data set. When both the ICD-10 codes and their corresponding laterality were considered, GPT-4V produced an average PPV of 7.8%, average TPR of 3.5%, and average F1 score of 4.5% on the NIH data set, and an average PPV of 10.9%, average TPR of 4.9%, and average F1 score of 6.4% on the MIDRC data set. With few-shot learning, GPT-4V showed improved performance on both data sets. When contrasting zero-shot and few-shot learning, there were improved average TPRs and F1 scores in the few-shot setting, but there was not a substantial increase in the average PPV. Conclusion Although GPT-4V has shown promise in understanding natural images, it had limited effectiveness in interpreting real-world chest radiographs. © RSNA, 2024 Supplemental material is available for this article.

Optic Nerve Diameter on Non-Contrast Computed Tomography and Intracranial Hypertension in Patients With Acute Brain Injury: A Validation Study.

Intracranial hypertension is a feared complication of acute brain injury that can cause ischemic stroke, herniation, and death. Identifying those at risk is difficult, and the physical examination is often confounded. Given the widespread availability and use of computed tomography (CT) in patients with acute brain injury, prior work has attempted to use optic nerve diameter measurements to identify those at risk of intracranial hypertension. We aimed to validate the use of optic nerve diameter measurements on CT as a screening tool for intracranial hypertension in a large cohort of brain-injured patients. We performed a retrospective observational cohort study in a single tertiary referral Neuroscience Intensive Care Unit. We identified patients with documented intracranial pressure (ICP) measures as part of their routine clinical care who had non-contrast CT head scans collected within 24 h, and then measured the optic nerve diameters and explored the relationship and test characteristics of these measures to identify those at risk of intracranial hypertension. In a cohort of 314 patients, optic nerve diameter on CT was linearly but weakly associated with ICP. When used to identify those with intracranial hypertension (> 20 mm Hg), the area under the receiver operator curve (AUROC) was 0.68. Using a previously proposed threshold of 0.6 cm, the sensitivity was 81%, specificity 43%, positive likelihood ratio 1.4, and negative likelihood ratio 0.45. CT-derived optic nerve diameter using a threshold of 0.6 cm is sensitive but not specific for intracranial hypertension, and the overall correlation is weak.

Comparing hematoma characteristics in primary intracerebral hemorrhage versus intracerebral hemorrhage caused by structural vascular lesions.

Intracerebral hemorrhage (ICH) caused by structural vascular lesions is associated with better outcomes than primary ICH, but this relationship is poorly understood. We tested the hypothesis that ICH from a vascular lesion has more benign hematoma characteristics compared to primary ICH. We performed a retrospective study using data from our medical center. The SMASH-U criteria were used to adjudicate the etiology of ICH. The co-primary outcomes were admission parenchymal hematoma volume and hematoma expansion at 24 h. Linear and logistic regression analyses were performed to test associations. A total of 231 patients were included of whom 42 (18%) had a vascular lesion. Compared to primary ICH patients, those with structural vascular lesions were younger (49 vs. 68 years, p < 0.001), less likely to have hypertension (29% vs. 74%, p < 0.001), had lower mean admission systolic blood pressure (140 ± 23 vs. 164 ± 35, p < 0.001), less frequently had IVH (26% vs. 44%, p = 0.03), and had mostly lobar or infratentorial hemorrhages. The median admission hematoma volume was smaller with vascular lesions (5.9 vs. 9.7 mL, p = 0.01). In regression models, ICH from a vascular lesion was associated with smaller admission hematoma volume (beta, -0.67, 95% CI, -1.29 to -0.05, p = 0.03), but no association with hematoma expansion was detected when assessed as a continuous (OR, 0.93; 95% CI, -4.46 to 6.30, p = 0.73) or dichotomous exposure (OR, 1.86; 95% CI, 0.40 to 8.51, p = 0.42). In a single-center cohort, patients with ICH from vascular lesions had smaller hematoma volumes than patients with primary ICH.

Ensembling Low Precision Models for Binary Biomedical Image Segmentation.

Segmentation of anatomical regions of interest such as vessels or small lesions in medical images is still a difficult problem that is often tackled with manual input by an expert. One of the major challenges for this task is that the appearance of foreground (positive) regions can be similar to background (negative) regions. As a result, many automatic segmentation algorithms tend to exhibit asymmetric errors, typically producing more false positives than false negatives. In this paper, we aim to leverage this asymmetry and train a diverse ensemble of models with very high recall, while sacrificing their precision. Our core idea is straightforward: A diverse ensemble of low precision and high recall models are likely to make different false positive errors (classifying background as foreground in different parts of the image), but the true positives will tend to be consistent. Thus, in aggregate the false positive errors will cancel out, yielding high performance for the ensemble. Our strategy is general and can be applied with any segmentation model. In three different applications (carotid artery segmentation in a neck CT angiography, myocardium segmentation in a cardiovascular MRI and multiple sclerosis lesion segmentation in a brain MRI), we show how the proposed approach can significantly boost the performance of a baseline segmentation method.

Differences in Admission Blood Pressure Among Causes of Intracerebral Hemorrhage.

Background and Purpose- It is unknown whether admission systolic blood pressure (SBP) differs among causes of intracerebral hemorrhage (ICH). We sought to elucidate an association between admission BP and ICH cause. Methods- We compared admission SBP across ICH causes among patients in the Cornell Acute Stroke Academic Registry, which includes all adults with ICH at our center from 2011 through 2017. Trained analysts prospectively collected demographics, comorbidities, and admission SBP, defined as the first recorded value in the emergency department or on transfer from another hospital. ICH cause was adjudicated by a panel of neurologists using the SMASH-U criteria. We used ANOVA to compare mean admission SBP among ICH causes. We used multiple linear regression to adjust for age, sex, race, Glasgow Coma Scale score, and hematoma size. In secondary analyses, we compared hourly SBP measurements during the first 72 hours after admission, using mixed-effects linear models adjusted for the covariates above plus antihypertensive agents. Results- Among 484 patients with ICH, admission SBP varied significantly across ICH causes, ranging from 138 (±24) mm Hg in those with structural vascular lesions to 167 (±35) mm Hg in those with hypertensive ICH (P<0.001). The mean admission SBP in hypertensive ICH was 17 (95% CI, 11-24) mm Hg higher than in ICH of all other causes combined. These differences remained significant after adjustment for age, sex, race, Glasgow Coma Scale score, and hematoma size (P<0.001), and this persisted throughout the first 72 hours of hospitalization (P<0.001). Conclusions- In a single-center ICH registry, SBP varied significantly among ICH causes, both on admission and during hospitalization. Our results suggest that BP in the acute post-ICH setting is at least partly associated with ICH cause rather than simply representing a physiological reaction to the ICH itself.