What question did this study set out to answer?

To assess the ability of deep learning models to classify sarcopenia severity based on MRI images in patients with diabetic foot disease.

March 29, 2026

SRS94 - Using Residual Network (ResNet) convolutional neural networks to classify sarcopenia severity on two-dimensional magnetic resonance images of the foot in patients with diabetic foot disease

Key Points

To assess the ability of deep learning models to classify sarcopenia severity based on MRI images in patients with diabetic foot disease.
Collected 1740 two-dimensional MRI images of the foot for analysis.
Classified images into mild, moderate, or severe sarcopenia.
Evaluated inter-rater reliability with a subset graded by a radiologist.
Used various ResNet models for classification tasks with training, validation, and testing splits.
Inter-rater reliability was confirmed at 0.827, indicating strong agreement.
ResNet-18 achieved 75% accuracy in classifying sarcopenia severity.
ResNet-34 reached 78.7% accuracy; ResNet-50 achieved 79% accuracy with an F1 score of 74%.
ResNet-101 improved accuracy to 80.7% with an F1 score of 69.3% in the same classification task.

Abstract

Abstract Introduction ResNet is a convolutional neural network (CNN) in the world of artificial intelligence and is used for image classification in the world of medicine. Initial ResNet structures (ResNet-18 and ResNet-34) had a basic CNN architecture whilst more recent models, ResNet-50 and ResNet-101 follow a bottleneck architecture. Diabetic foot disease (DFD) is a complex disease and is associated with lower limb amputation. Magnetic resonance imaging (MRI) is commonly used in patients with DFD. Aims To explore the use of deep learning models to identify predictors on MRI for limb loss in patients with DFD by using sarcoepenia as a potential surrogate marker. Methods Two-dimensional images of the foot at the base of the 1st metatarsal were classified as having mild, moderate or severe sarcopenia. A subset of 50 images were also graded by a musculoskeletal radiologist to establish the inter-rater reliability. 824 images were annotated. Following data pre-processing and data augmentation, 1740 images were available for the deep learning models which were split into a 70:20:10 ratio for training:validation:testing. Results The inter-rater reliability was 0.827 95% c.i. 0.726–0.928; P-value 0.001. ResNet-18 had a 75% accuracy in classifying sarcopenia severity, ResNet-34 had a 78.7% accuracy. ResNet-50 had an accuracy of 79% with an F1 score of 74% and ResNet-100 had an accuracy of 80.7% with an F1 score of 69.3% in classifying sarcopenia severity on two-dimensional MRI pictures. Conclusions ResNet models are useful in classifying the severity of sarcopenia in patients with diabetic foot disease.

Mark Helpful

Bookmark

Relay