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.

Standardized reporting for Head CT Scans in patients suspected of traumatic brain injury (TBI): An international expert endeavor.

BACKGROUND AND PURPOSE: Traumatic brain injury (TBI) is a major source of health loss and disability worldwide. Accurate and timely diagnosis of TBI is critical for appropriate treatment and management of the condition. Neuroimaging plays a crucial role in the diagnosis and characterization of TBI. Computed tomography (CT) is the first-line diagnostic imaging modality typically utilized in patients with suspected acute mild, moderate and severe TBI. Radiology reports play a crucial role in the diagnostic process, providing critical information about the location and extent of brain injury, as well as factors that could prevent secondary injury. However, the complexity and variability of radiology reports can make it challenging for healthcare providers to extract the necessary information for diagnosis and treatment planning. METHODS/RESULTS/CONCLUSION: In this article, we report the efforts of an international group of TBI imaging experts to develop a clinical radiology report template for CT scans obtained in patients suspected of TBI and consisting of fourteen different subdivisions (CT technique, mechanism of injury or clinical history, presence of scalp injuries, fractures, potential vascular injuries, potential injuries involving the extra-axial spaces, brain parenchymal injuries, potential injuries involving the cerebrospinal fluid spaces and the ventricular system, mass effect, secondary injuries, prior or coexisting pathology).

Association of partial T2-FLAIR mismatch sign and isocitrate dehydrogenase mutation in WHO grade 4 gliomas: results from the ReSPOND consortium.

PURPOSE: While the T2-FLAIR mismatch sign is highly specific for isocitrate dehydrogenase (IDH)-mutant, 1p/19q-noncodeleted astrocytomas among lower-grade gliomas, its utility in WHO grade 4 gliomas is not well-studied. We derived the partial T2-FLAIR mismatch sign as an imaging biomarker for IDH mutation in WHO grade 4 gliomas. METHODS: Preoperative MRI scans of adult WHO grade 4 glioma patients (n = 2165) from the multi-institutional ReSPOND (Radiomics Signatures for PrecisiON Diagnostics) consortium were analyzed. Diagnostic performance of the partial T2-FLAIR mismatch sign was evaluated. Subset analyses were performed to assess associations of imaging markers with overall survival (OS). RESULTS: One hundred twenty-one (5.6%) of 2165 grade 4 gliomas were IDH-mutant. Partial T2-FLAIR mismatch was present in 40 (1.8%) cases, 32 of which were IDH-mutant, yielding 26.4% sensitivity, 99.6% specificity, 80.0% positive predictive value, and 95.8% negative predictive value. Multivariate logistic regression demonstrated IDH mutation was significantly associated with partial T2-FLAIR mismatch (odds ratio [OR] 5.715, 95% CI [1.896, 17.221], p = 0.002), younger age (OR 0.911 [0.895, 0.927], p < 0.001), tumor centered in frontal lobe (OR 3.842, [2.361, 6.251], p < 0.001), absence of multicentricity (OR 0.173, [0.049, 0.612], p = 0.007), and presence of cystic (OR 6.596, [3.023, 14.391], p < 0.001) or non-enhancing solid components (OR 6.069, [3.371, 10.928], p < 0.001). Multivariate Cox analysis demonstrated cystic components (p = 0.024) and non-enhancing solid components (p = 0.003) were associated with longer OS, while older age (p < 0.001), frontal lobe center (p = 0.008), multifocality (p < 0.001), and multicentricity (p < 0.001) were associated with shorter OS. CONCLUSION: Partial T2-FLAIR mismatch sign is highly specific for IDH mutation in WHO grade 4 gliomas.

Radiomic signatures of meningiomas using the Ki-67 proliferation index as a prognostic marker of clinical outcomes.

OBJECTIVE: The clinical behavior of meningiomas is not entirely captured by its designated WHO grade, therefore other factors must be elucidated that portend increased tumor aggressiveness and associated risk of recurrence. In this study, the authors identify multiparametric MRI radiomic signatures of meningiomas using Ki-67 as a prognostic marker of clinical outcomes independent of WHO grade. METHODS: A retrospective analysis was conducted of all resected meningiomas between 2012 and 2018. Preoperative MR images were used for high-throughput radiomic feature extraction and subsequently used to develop a machine learning algorithm to stratify meningiomas based on Ki-67 indices < 5% and ≥ 5%, independent of WHO grade. Progression-free survival (PFS) was assessed based on machine learning prediction of Ki-67 strata and compared with outcomes based on histopathological Ki-67. RESULTS: Three hundred forty-three meningiomas were included: 291 with WHO grade I, 43 with grade II, and 9 with grade III. The overall rate of recurrence was 19.8% (15.1% in grade I, 44.2% in grade II, and 77.8% in grade III) over a median follow-up of 28.5 months. Grade II and III tumors had higher Ki-67 indices than grade I tumors, albeit tumor and peritumoral edema volumes had considerable variation independent of meningioma WHO grade. Forty-six high-performing radiomic features (1 morphological, 7 intensity-based, and 38 textural) were identified and used to build a support vector machine model to stratify tumors based on a Ki-67 cutoff of 5%, with resultant areas under the curve of 0.83 (95% CI 0.78-0.89) and 0.84 (95% CI 0.75-0.94) achieved for the discovery (n = 257) and validation (n = 86) data sets, respectively. Comparison of histopathological Ki-67 versus machine learning-predicted Ki-67 showed excellent performance (overall accuracy > 80%), with classification of grade I meningiomas exhibiting the greatest accuracy. Prediction of Ki-67 by machine learning classifier revealed shorter PFS for meningiomas with Ki-67 indices ≥ 5% compared with tumors with Ki-67 < 5% (p < 0.0001, log-rank test), which corroborates divergent patient outcomes observed using histopathological Ki-67. CONCLUSIONS: The Ki-67 proliferation index may serve as a surrogate marker of increased meningioma aggressiveness independent of WHO grade. Machine learning using radiomic feature analysis may be used for the preoperative prediction of meningioma Ki-67, which provides enhanced analytical insights to help improve diagnostic classification and guide patient-specific treatment strategies.

DTI of chronic spinal cord injury in children without MRI abnormalities (SCIWOMR) and with pathology on MRI and comparison to severity of motor impairment.

STUDY DESIGN: This investigation was a cohort study that included: 36 typically developing (TD) children and 19 children with spinal cord lesions who underwent spinal cord MRI. OBJECTIVES: To investigate diffusion tensor imaging (DTI) cervical and thoracic spinal cord changes in pediatric patients that have clinically traumatic and non-traumatic spinal cord injury (SCI) without MR (SCIWOMR) abnormalities. SETTING: Thomas Jefferson University, Temple University, Shriners Hospitals for Children all in Philadelphia, USA. METHODS: 36 TD children and 19 children with spinal cord lesions that represent either a chronic traumatic acquired SCI or chronic non-traumatic SCI (≥6 months post injury), age range, 6-16 years who underwent cervical and thoracic spinal cord MRI in 2014-2017. Additionally DTI was correlated to clinical American Spinal Injury Association Impairment Scale (AIS). RESULTS: Both SCIWOMR and MRI positive (+) groups showed abnormal FA and RD DTI values in the adjacent MRI-normal appearing segments of cephalad and caudal spinal cord compared to TD. The FA values demonstrated perilesional abnormal DTI findings in the middle and proximal segments of the cephalad and caudal cord in the SCIWOMR AIS A/B group compared to SCIWOMR AIS C/D group. CONCLUSIONS: We found DTI changes in children with SCIWOMR with different causes of spinal lesions. We also investigated the relationship between DTI and clinical AIS scores. This study further examined the potential diagnostic value of DTI and should be translatable to adults with spinal cord lesions.

Brain extraction on MRI scans in presence of diffuse glioma: Multi-institutional performance evaluation of deep learning methods and robust modality-agnostic training.

Brain extraction, or skull-stripping, is an essential pre-processing step in neuro-imaging that has a direct impact on the quality of all subsequent processing and analyses steps. It is also a key requirement in multi-institutional collaborations to comply with privacy-preserving regulations. Existing automated methods, including Deep Learning (DL) based methods that have obtained state-of-the-art results in recent years, have primarily targeted brain extraction without considering pathologically-affected brains. Accordingly, they perform sub-optimally when applied on magnetic resonance imaging (MRI) brain scans with apparent pathologies such as brain tumors. Furthermore, existing methods focus on using only T1-weighted MRI scans, even though multi-parametric MRI (mpMRI) scans are routinely acquired for patients with suspected brain tumors. In this study, we present a comprehensive performance evaluation of recent deep learning architectures for brain extraction, training models on mpMRI scans of pathologically-affected brains, with a particular focus on seeking a practically-applicable, low computational footprint approach, generalizable across multiple institutions, further facilitating collaborations. We identified a large retrospective multi-institutional dataset of n=3340 mpMRI brain tumor scans, with manually-inspected and approved gold-standard segmentations, acquired during standard clinical practice under varying acquisition protocols, both from private institutional data and public (TCIA) collections. To facilitate optimal utilization of rich mpMRI data, we further introduce and evaluate a novel ''modality-agnostic training'' technique that can be applied using any available modality, without need for model retraining. Our results indicate that the modality-agnostic approach1 obtains accurate results, providing a generic and practical tool for brain extraction on scans with brain tumors.

Histopathology-validated machine learning radiographic biomarker for noninvasive discrimination between true progression and pseudo-progression in glioblastoma.

BACKGROUND: Imaging of glioblastoma patients after maximal safe resection and chemoradiation commonly demonstrates new enhancements that raise concerns about tumor progression. However, in 30% to 50% of patients, these enhancements primarily represent the effects of treatment, or pseudo-progression (PsP). We hypothesize that quantitative machine learning analysis of clinically acquired multiparametric magnetic resonance imaging (mpMRI) can identify subvisual imaging characteristics to provide robust, noninvasive imaging signatures that can distinguish true progression (TP) from PsP. METHODS: We evaluated independent discovery (n = 40) and replication (n = 23) cohorts of glioblastoma patients who underwent second resection due to progressive radiographic changes suspicious for recurrence. Deep learning and conventional feature extraction methods were used to extract quantitative characteristics from the mpMRI scans. Multivariate analysis of these features revealed radiophenotypic signatures distinguishing among TP, PsP, and mixed response that compared with similar categories blindly defined by board-certified neuropathologists. Additionally, interinstitutional validation was performed on 20 new patients. RESULTS: Patients who demonstrate TP on neuropathology are significantly different (P < .0001) from those with PsP, showing imaging features reflecting higher angiogenesis, higher cellularity, and lower water concentration. The accuracy of the proposed signature in leave-one-out cross-validation was 87% for predicting PsP (area under the curve [AUC], 0.92) and 84% for predicting TP (AUC, 0.83), whereas in the discovery/replication cohort, the accuracy was 87% for predicting PsP (AUC, 0.84) and 78% for TP (AUC, 0.80). The accuracy in the interinstitutional cohort was 75% (AUC, 0.80). CONCLUSION: Quantitative mpMRI analysis via machine learning reveals distinctive noninvasive signatures of TP versus PsP after treatment of glioblastoma. Integration of the proposed method into clinical studies can be performed using the freely available Cancer Imaging Phenomics Toolkit.

The RSNA Pediatric Bone Age Machine Learning Challenge.

Purpose The Radiological Society of North America (RSNA) Pediatric Bone Age Machine Learning Challenge was created to show an application of machine learning (ML) and artificial intelligence (AI) in medical imaging, promote collaboration to catalyze AI model creation, and identify innovators in medical imaging. Materials and Methods The goal of this challenge was to solicit individuals and teams to create an algorithm or model using ML techniques that would accurately determine skeletal age in a curated data set of pediatric hand radiographs. The primary evaluation measure was the mean absolute distance (MAD) in months, which was calculated as the mean of the absolute values of the difference between the model estimates and those of the reference standard, bone age. Results A data set consisting of 14 236 hand radiographs (12 611 training set, 1425 validation set, 200 test set) was made available to registered challenge participants. A total of 260 individuals or teams registered on the Challenge website. A total of 105 submissions were uploaded from 48 unique users during the training, validation, and test phases. Almost all methods used deep neural network techniques based on one or more convolutional neural networks (CNNs). The best five results based on MAD were 4.2, 4.4, 4.4, 4.5, and 4.5 months, respectively. Conclusion The RSNA Pediatric Bone Age Machine Learning Challenge showed how a coordinated approach to solving a medical imaging problem can be successfully conducted. Future ML challenges will catalyze collaboration and development of ML tools and methods that can potentially improve diagnostic accuracy and patient care. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Siegel in this issue.

A Roadmap for Foundational Research on Artificial Intelligence in Medical Imaging: From the 2018 NIH/RSNA/ACR/The Academy Workshop.

Imaging research laboratories are rapidly creating machine learning systems that achieve expert human performance using open-source methods and tools. These artificial intelligence systems are being developed to improve medical image reconstruction, noise reduction, quality assurance, triage, segmentation, computer-aided detection, computer-aided classification, and radiogenomics. In August 2018, a meeting was held in Bethesda, Maryland, at the National Institutes of Health to discuss the current state of the art and knowledge gaps and to develop a roadmap for future research initiatives. Key research priorities include: 1, new image reconstruction methods that efficiently produce images suitable for human interpretation from source data; 2, automated image labeling and annotation methods, including information extraction from the imaging report, electronic phenotyping, and prospective structured image reporting; 3, new machine learning methods for clinical imaging data, such as tailored, pretrained model architectures, and federated machine learning methods; 4, machine learning methods that can explain the advice they provide to human users (so-called explainable artificial intelligence); and 5, validated methods for image de-identification and data sharing to facilitate wide availability of clinical imaging data sets. This research roadmap is intended to identify and prioritize these needs for academic research laboratories, funding agencies, professional societies, and industry.

Machine Learning in Neurooncology Imaging: From Study Request to Diagnosis and Treatment.

OBJECTIVE: Machine learning has potential to play a key role across a variety of medical imaging applications. This review seeks to elucidate the ways in which machine learning can aid and enhance diagnosis, treatment, and follow-up in neurooncology. CONCLUSION: Given the rapid pace of development in machine learning over the past several years, a basic proficiency of the key tenets and use cases in the field is critical to assessing potential opportunities and challenges of this exciting new technology.

Guidelines for the conduct of clinical trials in spinal cord injury: Neuroimaging biomarkers.

Traumatic spinal cord injury (SCI) leads to immediate neuronal and axonal damage at the focal injury site and triggers secondary pathologic series of events resulting in sensorimotor and autonomic dysfunction below the level of injury. Although there is no cure for SCI, neuroprotective and regenerative therapies show promising results at the preclinical stage. There is a pressing need to develop non-invasive outcome measures that can indicate whether a candidate therapeutic agent or a cocktail of therapeutic agents are positively altering the underlying disease processes. Recent conventional MRI studies have quantified spinal cord lesion characteristics and elucidated their relationship between severity of injury to clinical impairment and recovery. Next to the quantification of the primary cord damage, quantitative MRI measures of spinal cord (rostrocaudally to the lesion site) and brain integrity have demonstrated progressive and specific neurodegeneration of afferent and efferent neuronal pathways. MRI could therefore play a key role to ultimately uncover the relationship between clinical impairment/recovery and injury-induced neurodegenerative changes in the spinal cord and brain. Moreover, neuroimaging biomarkers hold promises to improve clinical trial design and efficiency through better patient stratification. The purpose of this narrative review is therefore to propose a guideline of clinically available MRI sequences and their derived neuroimaging biomarkers that have the potential to assess tissue damage at the macro- and microstructural level after SCI. In this piece, we make a recommendation for the use of key MRI sequences-both conventional and advanced-for clinical work-up and clinical trials.

Assessment of Critical Feeding Tube Malpositions on Radiographs Using Deep Learning.

Assess the efficacy of deep convolutional neural networks (DCNNs) in detection of critical enteric feeding tube malpositions on radiographs. 5475 de-identified HIPAA compliant frontal view chest and abdominal radiographs were obtained, consisting of 174 x-rays of bronchial insertions and 5301 non-critical radiographs, including normal course, normal chest, and normal abdominal x-rays. The ground-truth classification for enteric feeding tube placement was performed by two board-certified radiologists. Untrained and pretrained deep convolutional neural network models for Inception V3, ResNet50, and DenseNet 121 were each employed. The radiographs were fed into each deep convolutional neural network, which included untrained and pretrained models. The Tensorflow framework was used for Inception V3, ResNet50, and DenseNet. Images were split into training (4745), validation (630), and test (100). Both real-time and preprocessing image augmentation strategies were performed. Receiver operating characteristic (ROC) and area under the curve (AUC) on the test data were used to assess the models. Statistical differences among the AUCs were obtained. p < 0.05 was considered statistically significant. The pretrained Inception V3, which had an AUC of 0.87 (95 CI; 0.80-0.94), performed statistically significantly better (p < .001) than the untrained Inception V3, with an AUC of 0.60 (95 CI; 0.52-0.68). The pretrained Inception V3 also had the highest AUC overall, as compared with ResNet50 and DenseNet121, with AUC values ranging from 0.82 to 0.85. Each pretrained network outperformed its untrained counterpart. (p < 0.05). Deep learning demonstrates promise in differentiating critical vs. non-critical placement with an AUC of 0.87. Pretrained networks outperformed untrained ones in all cases. DCNNs may allow for more rapid identification and communication of critical feeding tube malpositions.

Comprehensive, Integrative Genomic Analysis of Diffuse Lower-Grade Gliomas.

BACKGROUND: Diffuse low-grade and intermediate-grade gliomas (which together make up the lower-grade gliomas, World Health Organization grades II and III) have highly variable clinical behavior that is not adequately predicted on the basis of histologic class. Some are indolent; others quickly progress to glioblastoma. The uncertainty is compounded by interobserver variability in histologic diagnosis. Mutations in IDH, TP53, and ATRX and codeletion of chromosome arms 1p and 19q (1p/19q codeletion) have been implicated as clinically relevant markers of lower-grade gliomas. METHODS: We performed genomewide analyses of 293 lower-grade gliomas from adults, incorporating exome sequence, DNA copy number, DNA methylation, messenger RNA expression, microRNA expression, and targeted protein expression. These data were integrated and tested for correlation with clinical outcomes. RESULTS: Unsupervised clustering of mutations and data from RNA, DNA-copy-number, and DNA-methylation platforms uncovered concordant classification of three robust, nonoverlapping, prognostically significant subtypes of lower-grade glioma that were captured more accurately by IDH, 1p/19q, and TP53 status than by histologic class. Patients who had lower-grade gliomas with an IDH mutation and 1p/19q codeletion had the most favorable clinical outcomes. Their gliomas harbored mutations in CIC, FUBP1, NOTCH1, and the TERT promoter. Nearly all lower-grade gliomas with IDH mutations and no 1p/19q codeletion had mutations in TP53 (94%) and ATRX inactivation (86%). The large majority of lower-grade gliomas without an IDH mutation had genomic aberrations and clinical behavior strikingly similar to those found in primary glioblastoma. CONCLUSIONS: The integration of genomewide data from multiple platforms delineated three molecular classes of lower-grade gliomas that were more concordant with IDH, 1p/19q, and TP53 status than with histologic class. Lower-grade gliomas with an IDH mutation either had 1p/19q codeletion or carried a TP53 mutation. Most lower-grade gliomas without an IDH mutation were molecularly and clinically similar to glioblastoma. (Funded by the National Institutes of Health.).

Leveraging Technology to Improve Radiology Workflow.

Although advancements in the last decade have automated much of the radiology workflow, there are several areas in the complex imaging process where standardization and innovation can be implemented. We discuss multiple tools and integrations that can help improve operational efficiency, quality, and safety.

Application of Color Transformation Techniques in Pediatric Spinal Cord MR Images: Typically Developing and Spinal Cord Injury Population.

The purpose of this study was to evaluate an improved and reliable visualization method for pediatric spinal cord MR images in healthy subjects and patients with spinal cord injury (SCI). A total of 15 pediatric volunteers (10 healthy subjects and 5 subjects with cervical SCI) with a mean age of 11.41 years (range 8-16 years) were recruited and scanned using a 3.0T Siemens Verio MR scanner. T2-weighted axial images were acquired covering entire cervical spinal cord level C1 to C7. These gray-scale images were then converted to color images by using five different techniques including hue-saturation-value (HSV), rainbow, red-green-blue (RGB), and two enhanced RGB techniques using automated contrast stretching and intensity inhomogeneity correction. Performance of these techniques was scored visually by two neuroradiologists within three selected cervical spinal cord intervertebral disk levels (C2-C3, C4-C5, and C6-C7) and quantified using signal to noise ratio (SNR) and contrast to noise ratio (CNR). Qualitative and quantitative evaluation of the color images shows consistent improvement across all the healthy and SCI subjects over conventional gray-scale T2-weighted gradient echo (GRE) images. Inter-observer reliability test showed moderate to strong intra-class correlation (ICC) coefficients in the proposed techniques (ICC > 0.73). The results suggest that the color images could be used for quantification and enhanced visualization of the spinal cord structures in addition to the conventional gray-scale images. This would immensely help towards improved delineation of the gray/white and CSF structures and further aid towards accurate manual or automatic drawings of region of interests (ROIs).