A usability design checklist for Mobile electronic data capturing forms: the validation process.

BACKGROUND: New Specific Application Domain (SAD) heuristics or design principles are being developed to guide the design and evaluation of mobile applications in a bid to improve on the usability of these applications. This is because the existing heuristics are rather generic and are often unable to reveal a large number of mobile usability issues related to mobile specific interfaces and characteristics. Mobile Electronic Data Capturing Forms (MEDCFs) are one of such applications that are being used to collect health data particularly in hard to reach areas, but with a number of usability challenges especially when used in rural areas by semi literate users. Existing SAD design principles are often not used to evaluate mobile forms because their focus on features specific to data capture is minimal. In addition, some of these lists are extremely long rendering them difficult to use during the design and development of the mobile forms. The main aim of this study therefore was to generate a usability evaluation checklist that can be used to design and evaluate Mobile Electronic Data Capturing Forms in a bid to improve their usability. We also sought to compare the novice and expert developers' views regarding usability criteria. METHODS: We conducted a literature review in August 2016 using key words on articles and gray literature, and those with a focus on heuristics for mobile applications, user interface designs of mobile devices and web forms were eligible for review. The data bases included the ACM digital library, IEEE-Xplore and Google scholar. We had a total of 242 papers after removing duplicates and a total of 10 articles which met the criteria were finally reviewed. This review resulted in an initial usability evaluation checklist consisting of 125 questions that could be adopted for designing MEDCFs. The questions that handled the five main categories in data capture namely; form content, form layout, input type, error handling and form submission were considered. A validation study was conducted with both novice and expert developers using a validation tool in a bid to refine the checklist which was based on 5 criteria. The criteria for the validation included utility, clarity, question naming, categorization and measurability, with utility and measurability having a higher weight respectively. We then determined the proportion of participants who agreed (scored 4 or 5), disagreed (scored 1 or 2) and were neutral (scored 3) to a given criteria regarding a particular question for each of the experts and novice developers. Finally, we selected questions that had an average of 85% agreement (scored 4 or 5) across all the 5 criteria by both novice and expert developers. 'Agreement' stands for capturing the same views or sentiments about the perceived likeness of an evaluation question. RESULTS: The validation study reduced the initial 125 usability evaluation questions to 30 evaluation questions with the form layout category having the majority questions. Results from the validation showed higher levels of affirmativeness from the expert developers compared to those of the novice developers across the different criteria; however the general trend of agreement on relevance of usability questions was similar across all the criteria for the developers. The evaluation questions that were being validated were found to be useful, clear, properly named and categorized, however the measurability of the questions was found not to be satisfactory by both sets of developers. The developers attached great importance to the use of appropriate language and to the visibility of the help function, but in addition expert developers felt that indication of mandatory and optional fields coupled with the use of device information like the Global Positioning System (GPS) was equally important. And for both sets of developers, utility had the highest scores while measurability scored least. CONCLUSION: The generated checklist indicated the design features the software developers found necessary to improve the usability of mobile electronic data collection tools. In the future, we thus propose to test the effectiveness of the measure for suitability and performance based on this generated checklist, and test it on the end users (data collectors) with a purpose of picking their design requirements. Continuous testing with the end users will help refine the checklist to include only that which is most important in improving the data collectors' experience.

Evaluating Molecular Properties Involved in Transport of Small Molecules in Stratum Corneum: A Quantitative Structure-Activity Relationship for Skin Permeability.

The skin permeability (Kp) defines the rate of a chemical penetrating across the stratum corneum. This value is widely used to quantitatively describe the transport of molecules in the outermost layer of epidermal skin and indicate the significance of skin absorption. This study defined a Kp quantitative structure-activity relationship (QSAR) based on 106 chemical substances of Kp measured using human skin and interpreted the molecular interactions underlying transport behavior of small molecules in the stratum corneum. The Kp QSAR developed in this study identified four molecular descriptors that described the molecular cyclicity in the molecule reflecting local geometrical environments, topological distances between pairs of oxygen and chlorine atoms, lipophilicity, and similarity to antineoplastics in molecular properties. This Kp QSAR considered the octanol-water partition coefficient to be a direct influence on transdermal movement of molecules. Moreover, the Kp QSAR identified a sub-domain of molecular properties initially defined to describe the antineoplastic resemblance of a compound as a significant factor in affecting transdermal permeation of solutes. This finding suggests that the influence of molecular size on the chemical's skin-permeating capability should be interpreted with other relevant physicochemical properties rather than being represented by molecular weight alone.

Accuracy of Radiomics in Predicting IDH Mutation Status in Diffuse Gliomas: A Bivariate Meta-Analysis.

Purpose To perform a systematic review and meta-analysis assessing the predictive accuracy of radiomics in the noninvasive determination of isocitrate dehydrogenase (IDH) status in grade 4 and lower-grade diffuse gliomas. Materials and Methods A systematic search was performed in the PubMed, Scopus, Embase, Web of Science, and Cochrane Library databases for relevant articles published between January 1, 2010, and July 7, 2021. Pooled sensitivity and specificity across studies were estimated. Risk of bias was evaluated using Quality Assessment of Diagnostic Accuracy Studies-2, and methods were evaluated using the radiomics quality score (RQS). Additional subgroup analyses were performed according to tumor grade, RQS, and number of sequences used (PROSPERO ID: CRD42021268958). Results Twenty-six studies that included 3280 patients were included for analysis. The pooled sensitivity and specificity of radiomics for the detection of IDH mutation were 79% (95% CI: 76, 83) and 80% (95% CI: 76, 83), respectively. Low RQS scores were found overall for the included works. Subgroup analyses showed lower false-positive rates in very low RQS studies (RQS < 6) (meta-regression, z = -1.9; P = .02) compared with adequate RQS studies. No substantial differences were found in pooled sensitivity and specificity for the pure grade 4 gliomas group compared with the all-grade gliomas group (81% and 86% vs 79% and 79%, respectively) and for studies using single versus multiple sequences (80% and 77% vs 79% and 82%, respectively). Conclusion The pooled data showed that radiomics achieved good accuracy performance in distinguishing IDH mutation status in patients with grade 4 and lower-grade diffuse gliomas. The overall methodologic quality (RQS) was low and introduced potential bias. Keywords: Neuro-Oncology, Radiomics, Integration, Application Domain, Glioblastoma, IDH Mutation, Radiomics Quality Scoring Supplemental material is available for this article. Published under a CC BY 4.0 license.

Vision Transformer-based Decision Support for Neurosurgical Intervention in Acute Traumatic Brain Injury: Automated Surgical Intervention Support Tool.

Purpose To develop an automated triage tool to predict neurosurgical intervention for patients with traumatic brain injury (TBI). Materials and Methods A provincial trauma registry was reviewed to retrospectively identify patients with TBI from 2005 to 2022 treated at a specialized Canadian trauma center. Model training, validation, and testing were performed using head CT scans with binary reference standard patient-level labels corresponding to whether the patient received neurosurgical intervention. Performance and accuracy of the model, the Automated Surgical Intervention Support Tool for TBI (ASIST-TBI), were also assessed using a held-out consecutive test set of all patients with TBI presenting to the center between March 2021 and September 2022. Results Head CT scans from 2806 patients with TBI (mean age, 57 years ± 22 [SD]; 1955 [70%] men) were acquired between 2005 and 2021 and used for training, validation, and testing. Consecutive scans from an additional 612 patients (mean age, 61 years ± 22; 443 [72%] men) were used to assess the performance of ASIST-TBI. There was accurate prediction of neurosurgical intervention with an area under the receiver operating characteristic curve (AUC) of 0.92 (95% CI: 0.88, 0.94), accuracy of 87% (491 of 562), sensitivity of 87% (196 of 225), and specificity of 88% (295 of 337) on the test dataset. Performance on the held-out test dataset remained robust with an AUC of 0.89 (95% CI: 0.85, 0.91), accuracy of 84% (517 of 612), sensitivity of 85% (199 of 235), and specificity of 84% (318 of 377). Conclusion A novel deep learning model was developed that could accurately predict the requirement for neurosurgical intervention using acute TBI CT scans. Keywords: CT, Brain/Brain Stem, Surgery, Trauma, Prognosis, Classification, Application Domain, Traumatic Brain Injury, Triage, Machine Learning, Decision Support Supplemental material is available for this article. © RSNA, 2024 See also commentary by Haller in this issue.

Development of predictive QSAR models for the substrates/inhibitors of OATP1B1 by deep neural networks.

The organic anion transporting polypeptide 1B1 (OATP1B1) is an important hepatic uptake transporter. Inhibition of its normal function could lead to drug-drug interactions. In silico prediction is an effective means to identify potential OATP1B1 inhibitors and quantitative structure-activity relationship (QSAR) modeling is extensively used. As the structures of OATP1B1 substrates/inhibitors are quite diverse, machine learning based methods should be a good option for their QSAR analysis. In the present study, deep neural networks (DNNs) were employed to develop QSAR models for the substrates/inhibitors of OATP1B1 with different molecular fingerprints. Our results showed that QSAR models based on 4-hidden layer DNNs and ECFP4/FCFP4 fingerprints had the best generalization performance. The correlation coefficients (R2) of test set for ECFP4 and FCFP4 models were 0.641 and 0.653, respectively. Model application domain (AD) was calculated with Euclidean distance-based method, and AD could improve the performance of ECFP4 model but has little effect on FCFP4 model. Finally, the prediction of additional 8 compounds that not included in the data set further demonstrated that our QSAR models had a good predictive ability (averaged prediction accuracy >92%). The developed QSAR models could be used to screen large data sets and discover novel inhibitors for OATP1B1.

Expansion of the application domain of a macromolecular ocular irritation test (OptiSafe™).

The OptiSafe (OS) test is shelf-stable, macromolecular eye irritation test that does not include any animal ingredient or component ("vegan"). The purpose of this study was to evaluate the test's accuracy for an expanded application domain for both the original and recently updated OS method. This study involved the testing of additional ocular corrosives and previously excluded foaming agents ("surfactants") using both the original and updated OS methods and then combining these data with prior validation data for a total of 147 chemicals. Predictivity was evaluated by a statistical comparison of the OptiSafe predictions with historical in vivo "Draize" rabbit eye data for the same chemicals (from public databases). We report that for the detection of chemicals not requiring classification for eye irritation [Globally Harmonized System of Classification and Labeling of Chemicals (GHS) No Category], the accuracy, specificity, and sensitivity were 92.8%, 79.6%, and 100.0%, respectively, for the updated method; for the detection of chemicals inducing extreme eye damage/corrosion (GHS Category 1), the accuracy, specificity, and sensitivity were 79.4%, 71.8%, and 91.7%, respectively, for the updated method. Results indicate that both the original and updated methods have a high accuracy for the expanded application domain that included ocular corrosives and surfactants.

Performance of a Chest Radiograph AI Diagnostic Tool for COVID-19: A Prospective Observational Study.

Purpose: To conduct a prospective observational study across 12 U.S. hospitals to evaluate real-time performance of an interpretable artificial intelligence (AI) model to detect COVID-19 on chest radiographs. Materials and Methods: A total of 95 363 chest radiographs were included in model training, external validation, and real-time validation. The model was deployed as a clinical decision support system, and performance was prospectively evaluated. There were 5335 total real-time predictions and a COVID-19 prevalence of 4.8% (258 of 5335). Model performance was assessed with use of receiver operating characteristic analysis, precision-recall curves, and F1 score. Logistic regression was used to evaluate the association of race and sex with AI model diagnostic accuracy. To compare model accuracy with the performance of board-certified radiologists, a third dataset of 1638 images was read independently by two radiologists. Results: Participants positive for COVID-19 had higher COVID-19 diagnostic scores than participants negative for COVID-19 (median, 0.1 [IQR, 0.0-0.8] vs 0.0 [IQR, 0.0-0.1], respectively; P < .001). Real-time model performance was unchanged over 19 weeks of implementation (area under the receiver operating characteristic curve, 0.70; 95% CI: 0.66, 0.73). Model sensitivity was higher in men than women (P = .01), whereas model specificity was higher in women (P = .001). Sensitivity was higher for Asian (P = .002) and Black (P = .046) participants compared with White participants. The COVID-19 AI diagnostic system had worse accuracy (63.5% correct) compared with radiologist predictions (radiologist 1 = 67.8% correct, radiologist 2 = 68.6% correct; McNemar P < .001 for both). Conclusion: AI-based tools have not yet reached full diagnostic potential for COVID-19 and underperform compared with radiologist prediction.Keywords: Diagnosis, Classification, Application Domain, Infection, Lung Supplemental material is available for this article.. © RSNA, 2022.

Utilization of Artificial Intelligence-based Intracranial Hemorrhage Detection on Emergent Noncontrast CT Images in Clinical Workflow.

Authors implemented an artificial intelligence (AI)-based detection tool for intracranial hemorrhage (ICH) on noncontrast CT images into an emergent workflow, evaluated its diagnostic performance, and assessed clinical workflow metrics compared with pre-AI implementation. The finalized radiology report constituted the ground truth for the analysis, and CT examinations (n = 4450) before and after implementation were retrieved using various keywords for ICH. Diagnostic performance was assessed, and mean values with their respective 95% CIs were reported to compare workflow metrics (report turnaround time, communication time of a finding, consultation time of another specialty, and turnaround time in the emergency department). Although practicable diagnostic performance was observed for overall ICH detection with 93.0% diagnostic accuracy, 87.2% sensitivity, and 97.8% negative predictive value, the tool yielded lower detection rates for specific subtypes of ICH (eg, 69.2% [74 of 107] for subdural hemorrhage and 77.4% [24 of 31] for acute subarachnoid hemorrhage). Common false-positive findings included postoperative and postischemic defects (23.6%, 37 of 157), artifacts (19.7%, 31 of 157), and tumors (15.3%, 24 of 157). Although workflow metrics such as communicating a critical finding (70 minutes [95% CI: 54, 85] vs 63 minutes [95% CI: 55, 71]) were on average reduced after implementation, future efforts are necessary to streamline the workflow all along the workflow chain. It is crucial to define a clear framework and recognize limitations as AI tools are only as reliable as the environment in which they are deployed. Keywords: CT, CNS, Stroke, Diagnosis, Classification, Application Domain © RSNA, 2022.

Classification of Multiple Diseases on Body CT Scans Using Weakly Supervised Deep Learning.

PURPOSE: To design multidisease classifiers for body CT scans for three different organ systems using automatically extracted labels from radiology text reports. MATERIALS AND METHODS: This retrospective study included a total of 12 092 patients (mean age, 57 years ± 18 [standard deviation]; 6172 women) for model development and testing. Rule-based algorithms were used to extract 19 225 disease labels from 13 667 body CT scans performed between 2012 and 2017. Using a three-dimensional DenseVNet, three organ systems were segmented: lungs and pleura, liver and gallbladder, and kidneys and ureters. For each organ system, a three-dimensional convolutional neural network classified each as no apparent disease or for the presence of four common diseases, for a total of 15 different labels across all three models. Testing was performed on a subset of 2158 CT volumes relative to 2875 manually derived reference labels from 2133 patients (mean age, 58 years ± 18; 1079 women). Performance was reported as area under the receiver operating characteristic curve (AUC), with 95% CIs calculated using the DeLong method. RESULTS: Manual validation of the extracted labels confirmed 91%-99% accuracy across the 15 different labels. AUCs for lungs and pleura labels were as follows: atelectasis, 0.77 (95% CI: 0.74, 0.81); nodule, 0.65 (95% CI: 0.61, 0.69); emphysema, 0.89 (95% CI: 0.86, 0.92); effusion, 0.97 (95% CI: 0.96, 0.98); and no apparent disease, 0.89 (95% CI: 0.87, 0.91). AUCs for liver and gallbladder were as follows: hepatobiliary calcification, 0.62 (95% CI: 0.56, 0.67); lesion, 0.73 (95% CI: 0.69, 0.77); dilation, 0.87 (95% CI: 0.84, 0.90); fatty, 0.89 (95% CI: 0.86, 0.92); and no apparent disease, 0.82 (95% CI: 0.78, 0.85). AUCs for kidneys and ureters were as follows: stone, 0.83 (95% CI: 0.79, 0.87); atrophy, 0.92 (95% CI: 0.89, 0.94); lesion, 0.68 (95% CI: 0.64, 0.72); cyst, 0.70 (95% CI: 0.66, 0.73); and no apparent disease, 0.79 (95% CI: 0.75, 0.83). CONCLUSION: Weakly supervised deep learning models were able to classify diverse diseases in multiple organ systems from CT scans.Keywords: CT, Diagnosis/Classification/Application Domain, Semisupervised Learning, Whole-Body Imaging© RSNA, 2022.

Automatic Diagnosis Labeling of Cardiovascular MRI by Using Semisupervised Natural Language Processing of Text Reports.

PURPOSE: To assess whether the semisupervised natural language processing (NLP) of text from clinical radiology reports could provide useful automated diagnosis categorization for ground truth labeling to overcome manual labeling bottlenecks in the machine learning pipeline. MATERIALS AND METHODS: In this retrospective study, 1503 text cardiac MRI reports from 2016 to 2019 were manually annotated for five diagnoses by clinicians: normal, dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy, myocardial infarction (MI), and myocarditis. A semisupervised method that uses bidirectional encoder representations from transformers (BERT) pretrained on 1.14 million scientific publications was fine-tuned by using the manually extracted labels, with a report dataset split into groups of 801 for training, 302 for validation, and 400 for testing. The model's performance was compared with two traditional NLP models: a rule-based model and a support vector machine (SVM) model. The models' F1 scores and receiver operating characteristic curves were used to analyze performance. RESULTS: After 15 epochs, the F1 scores on the test set of 400 reports were as follows: normal, 84%; DCM, 79%; hypertrophic cardiomyopathy, 86%; MI, 91%; and myocarditis, 86%. The pooled F1 score and area under the receiver operating curve were 86% and 0.96, respectively. On the same test set, the BERT model had a higher performance than the rule-based model (F1 score, 42%) and SVM model (F1 score, 82%). Diagnosis categories classified by using the BERT model performed the labeling of 1000 MR images in 0.2 second. CONCLUSION: The developed model used labels extracted from radiology reports to provide automated diagnosis categorization of MR images with a high level of performance.Keywords: Semisupervised Learning, Diagnosis/Classification/Application Domain, Named Entity Recognition, MRI Supplemental material is available for this article. © RSNA, 2021.

Overview of Noninterpretive Artificial Intelligence Models for Safety, Quality, Workflow, and Education Applications in Radiology Practice.

Artificial intelligence has become a ubiquitous term in radiology over the past several years, and much attention has been given to applications that aid radiologists in the detection of abnormalities and diagnosis of diseases. However, there are many potential applications related to radiologic image quality, safety, and workflow improvements that present equal, if not greater, value propositions to radiology practices, insurance companies, and hospital systems. This review focuses on six major categories for artificial intelligence applications: study selection and protocoling, image acquisition, worklist prioritization, study reporting, business applications, and resident education. All of these categories can substantially affect different aspects of radiology practices and workflows. Each of these categories has different value propositions in terms of whether they could be used to increase efficiency, improve patient safety, increase revenue, or save costs. Each application is covered in depth in the context of both current and future areas of work. Keywords: Use of AI in Education, Application Domain, Supervised Learning, Safety © RSNA, 2022.

Artificial Intelligence Detection of Missed Cancers at Digital Mammography That Were Detected at Digital Breast Tomosynthesis.

PURPOSE: To investigate how an artificial intelligence (AI) system performs at digital mammography (DM) from a screening population with ground truth defined by digital breast tomosynthesis (DBT), and whether AI could detect breast cancers at DM that had originally only been detected at DBT. MATERIALS AND METHODS: In this secondary analysis of data from a prospective study, DM examinations from 14 768 women (mean age, 57 years), examined with both DM and DBT with independent double reading in the MalmÓ§ Breast Tomosynthesis Screening Trial (MBTST) (ClinicalTrials.gov: NCT01091545; data collection, 2010-2015), were analyzed with an AI system. Of 136 screening-detected cancers, 95 cancers were detected at DM and 41 cancers were detected only at DBT. The system identifies suspicious areas in the image, scored 1-100, and provides a risk score of 1 to 10 for the whole examination. A cancer was defined as AI detected if the cancer lesion was correctly localized and scored at least 62 (threshold determined by the AI system developers), therefore resulting in the highest examination risk score of 10. Data were analyzed with descriptive statistics, and detection performance was analyzed with receiver operating characteristics. RESULTS: The highest examination risk score was assigned to 10% (1493 of 14 786) of the examinations. With 90.8% specificity, the AI system detected 75% (71 of 95) of the DM-detected cancers and 44% (18 of 41) of cancers at DM that had originally been detected only at DBT. The majority were invasive cancers (17 of 18). CONCLUSION: Almost half of the additional DBT-only screening-detected cancers in the MBTST were detected at DM with AI. AI did not reach double reading performance; however, if combined with double reading, AI has the potential to achieve a substantial portion of the benefit of DBT screening.Keywords: Computer-aided Diagnosis, Mammography, Breast, Diagnosis, Classification, Application DomainClinical trial registration no. NCT01091545© RSNA, 2021.

Automated Segmentation and Volume Measurement of Intracranial Internal Carotid Artery Calcification at Noncontrast CT.

PURPOSE: To develop and evaluate a fully-automated deep learning-based method for assessment of intracranial internal carotid artery calcification (ICAC). MATERIALS AND METHODS: This was a secondary analysis of prospectively collected data from the Rotterdam study (2003-2006) to develop and validate a deep learning-based method for automated ICAC delineation and volume measurement. Two observers manually delineated ICAC on noncontrast CT scans of 2319 participants (mean age, 69 years ± 7 [standard deviation]; 1154 women [53.2%]), and a deep learning model was trained to segment ICAC and quantify its volume. Model performance was assessed by comparing manual and automated segmentations and volume measurements to those produced by an independent observer (available on 47 scans), comparing the segmentation accuracy in a blinded qualitative visual comparison by an expert observer, and comparing the association with first stroke incidence from the scan date until 2016. All method performance metrics were computed using 10-fold cross-validation. RESULTS: The automated delineation of ICAC reached a sensitivity of 83.8% and positive predictive value (PPV) of 88%. The intraclass correlation between automatic and manual ICAC volume measures was 0.98 (95% CI: 0.97, 0.98; computed in the entire dataset). Measured between the assessments of independent observers, sensitivity was 73.9%, PPV was 89.5%, and intraclass correlation coefficient was 0.91 (95% CI: 0.84, 0.95; computed in the 47-scan subset). In the blinded visual comparisons of 294 regions, automated delineations were judged as more accurate than manual delineations in 131 regions, less accurate in 94 regions, and equally accurate in the rest of the regions (131 of 225, 58.2%; P = .01). The association of ICAC volume with incident stroke was similarly strong for both automated (hazard ratio, 1.38 [95% CI: 1.12, 1.75]) and manually measured volumes (hazard ratio, 1.48 [95% CI: 1.20, 1.87]). CONCLUSION: The developed model was capable of automated segmentation and volume quantification of ICAC with accuracy comparable to human experts.Keywords CT, Neural Networks, Carotid Arteries, Calcifications/Calculi, Arteriosclerosis, Segmentation, Vision Application Domain, Stroke Supplemental material is available for this article. © RSNA, 2021.

Development and Validation of a Convolutional Neural Network for Automated Detection of Scaphoid Fractures on Conventional Radiographs.

PURPOSE: To compare the performance of a convolutional neural network (CNN) to that of 11 radiologists in detecting scaphoid bone fractures on conventional radiographs of the hand, wrist, and scaphoid. MATERIALS AND METHODS: At two hospitals (hospitals A and B), three datasets consisting of conventional hand, wrist, and scaphoid radiographs were retrospectively retrieved: a dataset of 1039 radiographs (775 patients [mean age, 48 years ± 23 {standard deviation}; 505 female patients], period: 2017-2019, hospitals A and B) for developing a scaphoid segmentation CNN, a dataset of 3000 radiographs (1846 patients [mean age, 42 years ± 22; 937 female patients], period: 2003-2019, hospital B) for developing a scaphoid fracture detection CNN, and a dataset of 190 radiographs (190 patients [mean age, 43 years ± 20; 77 female patients], period: 2011-2020, hospital A) for testing the complete fracture detection system. Both CNNs were applied consecutively: The segmentation CNN localized the scaphoid and then passed the relevant region to the detection CNN for fracture detection. In an observer study, the performance of the system was compared with that of 11 radiologists. Evaluation metrics included the Dice similarity coefficient (DSC), Hausdorff distance (HD), sensitivity, specificity, positive predictive value (PPV), and area under the receiver operating characteristic curve (AUC). RESULTS: The segmentation CNN achieved a DSC of 97.4% ± 1.4 with an HD of 1.31 mm ± 1.03. The detection CNN had sensitivity of 78% (95% CI: 70, 86), specificity of 84% (95% CI: 77, 92), PPV of 83% (95% CI: 77, 90), and AUC of 0.87 (95% CI: 0.81, 0.91). There was no difference between the AUC of the CNN and that of the radiologists (0.87 [95% CI: 0.81, 0.91] vs 0.83 [radiologist range: 0.79-0.85]; P = .09). CONCLUSION: The developed CNN achieved radiologist-level performance in detecting scaphoid bone fractures on conventional radiographs of the hand, wrist, and scaphoid.Keywords: Convolutional Neural Network (CNN), Deep Learning Algorithms, Machine Learning Algorithms, Feature Detection-Vision-Application Domain, Computer-Aided DiagnosisSee also the commentary by Li and Torriani in this issue.Supplemental material is available for this article.©RSNA, 2021.

Synthesizing Quantitative T2 Maps in Right Lateral Knee Femoral Condyles from Multicontrast Anatomic Data with a Conditional Generative Adversarial Network.

PURPOSE: To develop a proof-of-concept convolutional neural network (CNN) to synthesize T2 maps in right lateral femoral condyle articular cartilage from anatomic MR images by using a conditional generative adversarial network (cGAN). MATERIALS AND METHODS: In this retrospective study, anatomic images (from turbo spin-echo and double-echo in steady-state scans) of the right knee of 4621 patients included in the 2004-2006 Osteoarthritis Initiative were used as input to a cGAN-based CNN, and a predicted CNN T2 was generated as output. These patients included men and women of all ethnicities, aged 45-79 years, with or at high risk for knee osteoarthritis incidence or progression who were recruited at four separate centers in the United States. These data were split into 3703 (80%) for training, 462 (10%) for validation, and 456 (10%) for testing. Linear regression analysis was performed between the multiecho spin-echo (MESE) and CNN T2 in the test dataset. A more detailed analysis was performed in 30 randomly selected patients by means of evaluation by two musculoskeletal radiologists and quantification of cartilage subregions. Radiologist assessments were compared by using two-sided t tests. RESULTS: The readers were moderately accurate in distinguishing CNN T2 from MESE T2, with one reader having random-chance categorization. CNN T2 values were correlated to the MESE values in the subregions of 30 patients and in the bulk analysis of all patients, with best-fit line slopes between 0.55 and 0.83. CONCLUSION: With use of a neural network-based cGAN approach, it is feasible to synthesize T2 maps in femoral cartilage from anatomic MRI sequences, giving good agreement with MESE scans.See also commentary by Yi and Fritz in this issue.Keywords: Cartilage Imaging, Knee, Experimental Investigations, Quantification, Vision, Application Domain, Convolutional Neural Network (CNN), Deep Learning Algorithms, Machine Learning Algorithms© RSNA, 2021.

MRI-based Identification and Classification of Major Intracranial Tumor Types by Using a 3D Convolutional Neural Network: A Retrospective Multi-institutional Analysis.

PURPOSE: To develop an algorithm to classify postcontrast T1-weighted MRI scans by tumor classes (high-grade glioma, low-grade glioma [LGG], brain metastasis, meningioma, pituitary adenoma, and acoustic neuroma) and a healthy tissue (HLTH) class. MATERIALS AND METHODS: In this retrospective study, preoperative postcontrast T1-weighted MR scans from four publicly available datasets-the Brain Tumor Image Segmentation dataset (n = 378), the LGG-1p19q dataset (n = 145), The Cancer Genome Atlas Glioblastoma Multiforme dataset (n = 141), and The Cancer Genome Atlas Low Grade Glioma dataset (n = 68)-and an internal clinical dataset (n = 1373) were used. In all, a total of 2105 images were split into a training dataset (n = 1396), an internal test set (n = 361), and an external test dataset (n = 348). A convolutional neural network was trained to classify the tumor type and to discriminate between images depicting HLTH and images depicting tumors. The performance of the model was evaluated by using cross-validation, internal testing, and external testing. Feature maps were plotted to visualize network attention. The accuracy, positive predictive value (PPV), negative predictive value, sensitivity, specificity, F1 score, area under the receiver operating characteristic curve (AUC), and area under the precision-recall curve (AUPRC) were calculated. RESULTS: On the internal test dataset, across the seven different classes, the sensitivities, PPVs, AUCs, and AUPRCs ranged from 87% to 100%, 85% to 100%, 0.98 to 1.00, and 0.91 to 1.00, respectively. On the external data, they ranged from 91% to 97%, 73% to 99%, 0.97 to 0.98, and 0.9 to 1.0, respectively. CONCLUSION: The developed model was capable of classifying postcontrast T1-weighted MRI scans of different intracranial tumor types and discriminating images depicting pathologic conditions from images depicting HLTH.Keywords MR-Imaging, CNS, Brain/Brain Stem, Diagnosis/Classification/Application Domain, Supervised Learning, Convolutional Neural Network, Deep Learning Algorithms, Machine Learning Algorithms Supplemental material is available for this article. © RSNA, 2021.

Automated Morphometric Analysis of the Hip Joint on MRI from the German National Cohort Study.

PURPOSE: To develop and validate an automated morphometric analysis framework for the quantitative analysis of geometric hip joint parameters in MR images from the German National Cohort (GNC) study. MATERIALS AND METHODS: A secondary analysis on 40 participants (mean age, 51 years; age range, 30-67 years; 25 women) from the prospective GNC MRI study (2015-2016) was performed. Based on a proton density-weighted three-dimensional fast spin-echo sequence, a morphometric analysis approach was developed, including deep learning-based landmark localization, bone segmentation of the femora and pelvis, and a shape model for annotation transfer. The centrum-collum-diaphyseal, center-edge (CE), three alpha angles, head-neck offset (HNO), and HNO ratio along with the acetabular depth, inclination, and anteversion were derived. Quantitative validation was provided by comparison with average manual assessments of radiologists in a cross-validation format. Paired-sample t tests with a Bonferroni-corrected significance level of .005 were employed alongside mean differences and 10th/90th percentiles, median absolute deviations (MADs), and intraclass correlation coefficients (ICCs). RESULTS: High agreement in mean Dice similarity coefficients was achieved (average of 97.52% ± 0.46 [standard deviation]). The subsequent morphometric analysis produced results with low mean MAD values, with the highest values of 3.34° (alpha 03:00 o'clock position) and 0.87 mm (HNO) and ICC values ranging between 0.288 (HNO ratio) and 0.858 (CE) compared with manual assessments. These values were in line with interreader agreements, which at most had MAD values of 4.02° (alpha 12:00 o'clock position) and 1.07 mm (HNO) and ICC values ranging between 0.218 (HNO ratio) and 0.777 (CE). CONCLUSION: Automatic extraction of geometric hip parameters from MRI is feasible using a morphometric analysis approach with deep learning.Keywords: Computer-Aided Diagnosis (CAD), Interventional-MSK, MR-Imaging, Neural Networks, Skeletal-Appendicular, Hip, Anatomy, Computer Applications-3D, Segmentation, Vision, Application Domain, Quantification Supplemental material is available for this article. © RSNA, 2021.

Bioactivity Prediction Based on Matched Molecular Pair and Matched Molecular Series Methods.

BACKGROUND: Enhancing a compound's biological activity is the central task for lead optimization in small molecules drug discovery. However, it is laborious to perform many iterative rounds of compound synthesis and bioactivity tests. To address the issue, it is highly demanding to develop high quality in silico bioactivity prediction approaches, to prioritize such more active compound derivatives and reduce the trial-and-error process. METHODS: Two kinds of bioactivity prediction models based on a large-scale structure-activity relationship (SAR) database were constructed. The first one is based on the similarity of substituents and realized by matched molecular pair analysis, including SA, SA_BR, SR, and SR_BR. The second one is based on SAR transferability and realized by matched molecular series analysis, including Single MMS pair, Full MMS series, and Multi single MMS pairs. Moreover, we also defined the application domain of models by using the distance-based threshold. RESULTS: Among seven individual models, Multi single MMS pairs bioactivity prediction model showed the best performance (R2 = 0.828, MAE = 0.406, RMSE = 0.591), and the baseline model (SA) produced the most lower prediction accuracy (R2 = 0.798, MAE = 0.446, RMSE = 0.637). The predictive accuracy could further be improved by consensus modeling (R2 = 0.842, MAE = 0.397 and RMSE = 0.563). CONCLUSION: An accurate prediction model for bioactivity was built with a consensus method, which was superior to all individual models. Our model should be a valuable tool for lead optimization.

Comparison of prediction methods for octanol-air partition coefficients of diverse organic compounds.

The octanol-air partition coefficient (KOA) is needed for assessing multimedia transport and bioaccumulability of organic chemicals in the environment. As experimental determination of KOA for various chemicals is costly and laborious, development of KOA estimation methods is necessary. We investigated three methods for KOA prediction, conventional quantitative structure-activity relationship (QSAR) models based on molecular structural descriptors, group contribution models based on atom-centered fragments, and a novel model that predicts KOA via solvation free energy from air to octanol phase (ΔGO(0)), with a collection of 939 experimental KOA values for 379 compounds at different temperatures (263.15-323.15 K) as validation or training sets. The developed models were evaluated with the OECD guidelines on QSAR models validation and applicability domain (AD) description. Results showed that although the ΔGO(0) model is theoretically sound and has a broad AD, the prediction accuracy of the model is the poorest. The QSAR models perform better than the group contribution models, and have similar predictability and accuracy with the conventional method that estimates KOA from the octanol-water partition coefficient and Henry's law constant. One QSAR model, which can predict KOA at different temperatures, was recommended for application as to assess the long-range transport potential of chemicals.