Potential of GPT-4 for Detecting Errors in Radiology Reports: Implications for Reporting Accuracy.

Background Errors in radiology reports may occur because of resident-to-attending discrepancies, speech recognition inaccuracies, and large workload. Large language models, such as GPT-4 (ChatGPT; OpenAI), may assist in generating reports. Purpose To assess effectiveness of GPT-4 in identifying common errors in radiology reports, focusing on performance, time, and cost-efficiency. Materials and Methods In this retrospective study, 200 radiology reports (radiography and cross-sectional imaging [CT and MRI]) were compiled between June 2023 and December 2023 at one institution. There were 150 errors from five common error categories (omission, insertion, spelling, side confusion, and other) intentionally inserted into 100 of the reports and used as the reference standard. Six radiologists (two senior radiologists, two attending physicians, and two residents) and GPT-4 were tasked with detecting these errors. Overall error detection performance, error detection in the five error categories, and reading time were assessed using Wald χ2 tests and paired-sample t tests. Results GPT-4 (detection rate, 82.7%;124 of 150; 95% CI: 75.8, 87.9) matched the average detection performance of radiologists independent of their experience (senior radiologists, 89.3% [134 of 150; 95% CI: 83.4, 93.3]; attending physicians, 80.0% [120 of 150; 95% CI: 72.9, 85.6]; residents, 80.0% [120 of 150; 95% CI: 72.9, 85.6]; P value range, .522-.99). One senior radiologist outperformed GPT-4 (detection rate, 94.7%; 142 of 150; 95% CI: 89.8, 97.3; P = .006). GPT-4 required less processing time per radiology report than the fastest human reader in the study (mean reading time, 3.5 seconds ± 0.5 [SD] vs 25.1 seconds ± 20.1, respectively; P < .001; Cohen d = -1.08). The use of GPT-4 resulted in lower mean correction cost per report than the most cost-efficient radiologist ($0.03 ± 0.01 vs $0.42 ± 0.41; P < .001; Cohen d = -1.12). Conclusion The radiology report error detection rate of GPT-4 was comparable with that of radiologists, potentially reducing work hours and cost. © RSNA, 2024 See also the editorial by Forman in this issue.

Five-minute knee MRI: An AI-based super resolution reconstruction approach for compressed sensing. A validation study on healthy volunteers.

PURPOSE: To investigate the potential of combining Compressed Sensing (CS) and a newly developed AI-based super resolution reconstruction prototype consisting of a series of convolutional neural networks (CNN) for a complete five-minute 2D knee MRI protocol. METHODS: In this prospective study, 20 volunteers were examined using a 3T-MRI-scanner (Ingenia Elition X, Philips). Similar to clinical practice, the protocol consists of a fat-saturated 2D-proton-density-sequence in coronal, sagittal and transversal orientation as well as a sagittal T1-weighted sequence. The sequences were acquired with two different resolutions (standard and low resolution) and the raw data reconstructed with two different reconstruction algorithms: a conventional Compressed SENSE (CS) and a new CNN-based algorithm for denoising and subsequently to interpolate and therewith increase the sharpness of the image (CS-SuperRes). Subjective image quality was evaluated by two blinded radiologists reviewing 8 criteria on a 5-point Likert scale and signal-to-noise ratio calculated as an objective parameter. RESULTS: The protocol reconstructed with CS-SuperRes received higher ratings than the time-equivalent CS reconstructions, statistically significant especially for low resolution acquisitions (e.g., overall image impression: 4.3 ±â€¯0.4 vs. 3.4 ±â€¯0.4, p < 0.05). CS-SuperRes reconstructions for the low resolution acquisition were comparable to traditional CS reconstructions with standard resolution for all parameters, achieving a scan time reduction from 11:01 min to 4:46 min (57 %) for the complete protocol (e.g. overall image impression: 4.3 ±â€¯0.4 vs. 4.0 ±â€¯0.5, p < 0.05). CONCLUSION: The newly-developed AI-based reconstruction algorithm CS-SuperRes allows to reduce scan time by 57% while maintaining unchanged image quality compared to the conventional CS reconstruction.

In vivo characterization of a podocyte-expressed short podocin isoform.

The most common genetic causes of steroid-resistant nephrotic syndrome (SRNS) are mutations in the NPHS2 gene, which encodes the cholesterol-binding, lipid-raft associated protein podocin. Mass spectrometry and cDNA sequencing revealed the existence of a second shorter isoform in the human kidney in addition to the well-studied canonical full-length protein. Distinct subcellular localization of the shorter isoform that lacks part of the conserved PHB domain suggested a physiological role. Here, we analyzed whether this protein can substitute for the canonical full-length protein. The short isoform of podocin is not found in other organisms except humans. We therefore analysed a mouse line expressing the equivalent podocin isoform (podocinΔexon5) by CRISPR/Cas-mediated genome editing. We characterized the phenotype of these mice expressing podocinΔexon5 and used targeted mass spectrometry and qPCR to compare protein and mRNA levels of podocinwildtype and podocinΔexon5. After immunolabeling slit diaphragm components, STED microscopy was applied to visualize alterations of the podocytes' foot process morphology.Mice homozygous for podocinΔexon5 were born heavily albuminuric and did not survive past the first 24 h after birth. Targeted mass spectrometry revealed massively decreased protein levels of podocinΔexon5, whereas mRNA abundance was not different from the canonical form of podocin. STED microscopy revealed the complete absence of podocin at the podocytes' slit diaphragm and severe morphological alterations of podocyte foot processes. Mice heterozygous for podocinΔexon5 were phenotypically and morphologically unaffected despite decreased podocin and nephrin protein levels.The murine equivalent to the human short isoform of podocin cannot stabilize the lipid-protein complex at the podocyte slit diaphragm. Reduction of podocin levels at the site of the slit diaphragm complex has a detrimental effect on podocyte function and morphology. It is associated with decreased protein abundance of nephrin, the central component of the filtration-slit forming slit diaphragm protein complex.

Reconstruction of shoulder MRI using deep learning and compressed sensing: a validation study on healthy volunteers.

BACKGROUND: To investigate the potential of combining compressed sensing (CS) and deep learning (DL) for accelerated two-dimensional (2D) and three-dimensional (3D) magnetic resonance imaging (MRI) of the shoulder. METHODS: Twenty healthy volunteers were examined using at 3-T scanner with a fat-saturated, coronal, 2D proton density-weighted sequence with four acceleration levels (2.3, 4, 6, and 8) and a 3D sequence with three acceleration levels (8, 10, and 13), all accelerated with CS and reconstructed using the conventional algorithm and a new DL-based algorithm (CS-AI). Subjective image quality was evaluated by two blinded readers using 6 criteria on a 5-point Likert scale (overall impression, artifacts, and delineation of the subscapularis tendon, bone, acromioclavicular joint, and glenoid labrum). Objective image quality was measured by calculating signal-to-noise-ratio, contrast-to-noise-ratio, and a structural similarity index measure. All reconstructions were compared to the clinical standard (CS 2D acceleration factor 2.3; CS 3D acceleration factor 8). Additionally, subjective and objective image quality were compared between CS and CS-AI with the same acceleration levels. RESULTS: Both 2D and 3D sequences reconstructed with CS-AI achieved on average significantly better subjective and objective image quality compared to sequences reconstructed with CS with the same acceleration factor (p ≤ 0.011). Comparing CS-AI to the reference sequences showed that 4-fold acceleration for 2D sequences and 13-fold acceleration for 3D sequences without significant loss of quality (p ≥ 0.058). CONCLUSIONS: For MRI of the shoulder at 3 T, a DL-based algorithm allowed additional acceleration of acquisition times compared to the conventional approach. RELEVANCE STATEMENT: The combination of deep-learning and compressed sensing hold the potential for further scan time reduction in 2D and 3D imaging of the shoulder while providing overall better objective and subjective image quality compared to the conventional approach. TRIAL REGISTRATION: DRKS00024156. KEY POINTS: • Combination of compressed sensing and deep learning improved image quality and allows for significant acceleration of shoulder MRI. • Deep learning-based algorithm achieved better subjective and objective image quality than conventional compressed sensing. • For shoulder MRI at 3 T, 40% faster image acquisition for 2D sequences and 38% faster image acquisition for 3D sequences may be possible.

Super-Resolution Imaging of the Filtration Barrier Suggests a Role for Podocin R229Q in Genetic Predisposition to Glomerular Disease.

BACKGROUND: Diseases of the kidney's glomerular filtration barrier are a leading cause of end stage renal failure. Despite a growing understanding of genes involved in glomerular disorders in children, the vast majority of adult patients lack a clear genetic diagnosis. The protein podocin p.R229Q, which results from the most common missense variant in NPHS2, is enriched in cohorts of patients with FSGS. However, p.R229Q has been proposed to cause disease only when transassociated with specific additional genetic alterations, and population-based epidemiologic studies on its association with albuminuria yielded ambiguous results. METHODS: To test whether podocin p.R229Q may also predispose to the complex disease pathogenesis in adults, we introduced the exact genetic alteration in mice using CRISPR/Cas9-based genome editing (PodR231Q ). We assessed the phenotype using super-resolution microscopy and albuminuria measurements and evaluated the stability of the mutant protein in cell culture experiments. RESULTS: Heterozygous PodR231Q/wild-type mice did not present any overt kidney disease or proteinuria. However, homozygous PodR231Q/R231Q mice developed increased levels of albuminuria with age, and super-resolution microscopy revealed preceding ultrastructural morphologic alterations that were recently linked to disease predisposition. When injected with nephrotoxic serum to induce glomerular injury, heterozygous PodR231Q/wild-type mice showed a more severe course of disease compared with Podwild-type/wild-type mice. Podocin protein levels were decreased in PodR231Q/wild-type and PodR231Q/R231Q mice as well as in human cultured podocytes expressing the podocinR231Q variant. Our in vitro experiments indicate an underlying increased proteasomal degradation. CONCLUSIONS: Our findings demonstrate that podocin R231Q exerts a pathogenic effect on its own, supporting the concept of podocin R229Q contributing to genetic predisposition in adult patients.

A molecular mechanism explaining albuminuria in kidney disease.

Mammalian kidneys constantly filter large amounts of liquid, with almost complete retention of albumin and other macromolecules in the plasma. Breakdown of the three-layered renal filtration barrier results in loss of albumin into urine (albuminuria) across the wall of small renal capillaries, and is a leading cause of chronic kidney disease. However, exactly how the renal filter works and why its permeability is altered in kidney diseases is poorly understood. Here we show that the permeability of the renal filter is modulated through compression of the capillary wall. We collect morphometric data prior to and after onset of albuminuria in a mouse model equivalent to a human genetic disease affecting the renal filtration barrier. Combining quantitative analyses with mathematical modelling, we demonstrate that morphological alterations of the glomerular filtration barrier lead to reduced compressive forces that counteract filtration pressure, thereby resulting in capillary dilatation, and ultimately albuminuria. Our results reveal distinct functions of the different layers of the filtration barrier and expand the molecular understanding of defective renal filtration in chronic kidney disease.