Effect of Patient Clinical Variables in Osteoporosis Classification Using Hip X-rays in Deep Learning Analysis

Abstract

Background and Objectives: A few deep learning studies have reported that combining image features with patient variables enhanced identification accuracy compared with image-only models. However, previous studies have not statistically reported the additional effect of patient variables on the image-only models. This study aimed to statistically evaluate the osteoporosis identification ability of deep learning by combining hip radiographs with patient variables. Materials andMethods: We collected a dataset containing 1699 images from patients who underwent skeletal-bone-mineral density measurements and hip radiography at a general hospital from 2014 to 2021. Osteoporosis was assessed from hip radiographs using convolutional neural network (CNN) models (ResNet18, 34, 50, 101, and 152). We also investigated ensemble models with patient clinical variables added to each CNN. Accuracy, precision, recall, specificity, F1 score, and area under the curve (AUC) were calculated as performance metrics. Furthermore, we statistically compared the accuracy of the image-only model with that of an ensemble model that included images plus patient factors, including effect size for each performance metric. Results: All metrics were improved in the ResNet34 ensemble model compared with the image-only model. The AUC score in the ensemble model was significantly improved compared with the image-only model (difference 0.004; 95% CI 0.002–0.0007; p = 0.0004, effect size: 0.871). Conclusions: This study revealed the additional effect of patient variables in identification of osteoporosis using deep CNNs with hip radiographs. Our results provided evidence that the patient variables had additive synergistic effects on the image in osteoporosis identification.

Artificial Intelligence in Orthopedic Radiography Analysis: A Narrative Review

“…They tested five different CNNs and found high accuracy. When adding routinely available patient variables such as age, gender, and BMI, the diagnostic accuracy improved to an AUC of 0.906 with a ResNet50 CNN architecture [ 46 , 47 ].…”

Is attention branch network effective in classifying dental implants from panoramic radiograph images by deep learning?

“…However, increasing the number of parameters in the deep layers leads to the burden of calculation cost. By adding another structure to the CNN structure [ 15 ] or changing the structure of the CNNs [ 16 ], various developments have been made, such as achieving the same accuracy with a small number of parameters.…”

Effective deep learning for oral exfoliative cytology classification

“…Deep learning using CNN has had a great effect on the classification of medical images 9 , 10 . The development of various deep learning CNN models 11 , 12 and various optimization algorithms to improve the classification accuracy is rapidly progressing. There are various optimization algorithms, and in recent years, Sharpness Aware Minimization (SAM) 13 has been reported as an effective learning method for CNNs.…”