Abstract
Develop a highly accurate deep learning model to reliably classify radiographs by laterality. Digital Imaging and Communications in Medicine (DICOM) data for nine body parts was extracted retrospectively. Laterality was determined directly if encoded properly or inferred using other elements. Curation confirmed categorization and identified inaccurate labels due to human error. Augmentation enriched training data to semi-equilibrate classes. Classification and object detection models were developed on a dedicated workstation and tested on novel images. Receiver operating characteristic (ROC) curves, sensitivity, specificity, and accuracy were calculated. Study-level accuracy was determined and both were compared to human performance. An ensemble model was tested for the rigorous use-case of automatically classifying exams retrospectively. The final classification model identified novel images with an ROC area under the curve (AUC) of 0.999, improving on previous work and comparable to human performance. A similar ROC curve was observed for per-study analysis with AUC of 0.999. The object detection model classified images with accuracy of 99% or greater at both image and study level. Confidence scores allow adjustment of sensitivity and specificity as needed; the ensemble model designed for the highly specific use-case of automatically classifying exams was comparable and arguably better than human performance demonstrating 99% accuracy with 1% of exams unchanged and no incorrect classification. Deep learning models can classify radiographs by laterality with high accuracy and may be applied in a variety of settings that could improve patient safety and radiologist satisfaction. Rigorous use-cases requiring high specificity are achievable.
Highlights
Machine learning (ML) and deep learning (DL) are artificial intelligence (AI) methods with significant potential to augment both interpretive and non-interpretive radiology workflows
The 15,405 unique images curated for training, validation, and testing came from 4619 unique exams, of which 694 (15%)
Sixty-seven images from 23 exams had frankly incorrect Digital Imaging and Communications in Medicine (DICOM) laterality data compared to the lead marker
Summary
Machine learning (ML) and deep learning (DL) are artificial intelligence (AI) methods with significant potential to augment both interpretive and non-interpretive radiology workflows. Ongoing research displays promise in diverse applications of diagnosis, enhanced imaging and reconstruction, automated decision support, exam prioritization, and risk prediction [1,2,3,4,5,6]. ML involves the application of mathematical models to datasets to generate autonomous predictions using new data. Exposure to training data allows the model to learn from errors in processing initial cases with iterative improvement in performance after additional examples. A common ML algorithm is the artificial neural network (ANN) consisting of three segments (input, hidden, and output) of which many hidden layers can exist [1]. In order to train these networks reliably, large accurately labeled and curated datasets are required [2]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
