Logo

American Heart Association

  23
  0


Final ID: MP2574

Deep Learning-Guided CT Image Analysis Quantifies 18-Month Changes in Regional Muscle Atrophy in Patients with Peripheral Artery Disease

Abstract Body (Do not enter title and authors here): Introduction: Peripheral artery disease (PAD) is characterized by atherosclerosis of lower extremity arteries that promotes reduced muscle perfusion and skeletal muscle atrophy, which contributes to functional impairment in PAD patients. Computed tomography (CT) imaging offers a non-invasive method for quantifying peripheral muscle characteristics; however, manual image segmentation of muscles is time-consuming, and regional muscle analysis remains understudied in PAD. Therefore, we sought to develop and validate a deep learning approach for calf muscle segmentation to assess serial regional changes in muscle density of PAD patients over an 18-month study period.

Methods: Patients with PAD (n=89) were prospectively enrolled for lower extremity non-contrast CT imaging. A subset of patients (n=45) was recruited for an additional follow-up CT scan 18 months later. Calf muscle groups, including the gastrocnemius, soleus, and tibialis anterior, were manually segmented from axial CT images of each patient’s symptomatic limb, and mean muscle densities for each muscle were quantified based on CT Hounsfield units. Manual segmentations were used to train an nnU-Net deep learning model. Data augmentation techniques were performed to enhance generalization. The dataset was split into an 80/20 ratio for the training and test sets. Dice coefficients were calculated to evaluate the overlap/agreement between manual and deep learning segmentations. Paired t-tests were performed to assess the differences in muscle densities between the baseline and 18-month follow-up measures.

Results: The deep learning model achieved high segmentation performance, with dice scores of 0.90 ± 0.02 for the gastrocnemius and soleus, and 0.89 ± 0.02 for the tibialis anterior. Deep learning-guided serial CT image analysis detected a significant reduction in muscle densities across all muscle groups at 18-month follow-up when compared to baseline measurements (p<0.05).

Conclusion: Deep learning analysis can enable rapid, regional analysis of lower extremity skeletal muscle characteristics, reducing the time for muscle-by-muscle analysis in PAD patients from hours to seconds. Future implementation of regional muscle analysis in PAD may assist vascular medicine specialists with identifying regional muscle wasting that is related to patient walking impairment and/or symptoms as well as aid in serial monitoring of PAD patients following revascularization or supervised exercise for claudication.
  • Musini, Kumudha  ( Ohio State University , Columbus , Ohio , United States )
  • Rimmerman, Eleanor  ( The Ohio State University , Columbus , Ohio , United States )
  • Chou, Ting-heng  ( NTUNHS , Taipei , Taiwan )
  • Shin, Kyle  ( OSUCOM , Columbus , Ohio , United States )
  • Go, Michael  ( Ohio State College of Medicine , Columbus , Ohio , United States )
  • Stacy, Mitchel  ( Ohio State College of Medicine , Columbus , Ohio , United States )
  • Author Disclosures:
    Kumudha Musini: DO NOT have relevant financial relationships | Eleanor Rimmerman: DO NOT have relevant financial relationships | Ting-Heng Chou: DO NOT have relevant financial relationships | Kyle Shin: DO NOT have relevant financial relationships | Michael Go: No Answer | Mitchel Stacy: DO NOT have relevant financial relationships
Meeting Info:

Scientific Sessions 2025

2025

New Orleans, Louisiana

Session Info:

Best of Vascular Imaging

Monday, 11/10/2025 , 10:45AM - 12:00PM

Moderated Digital Poster Session

More abstracts on this topic:
AI-driven Measurement of Myosteatosis in Coronary Artery Calcium Scans Predicts Atrial Fibrillation and Heart Failure. An AI-CVD Study within the Multi-Ethnic Study of Atherosclerosis (MESA)

Naghavi Morteza, Wong Nathan, Vliegenthart Rozemarijn, Mcconnell Michael, Maron David, Branch Andrea, Kloner Robert, Mechanick Jeffrey, Ma Ning, Yip Rowena, Fan Wenjun, Azimi Amir, Roy Sion, Nasir Khurram, Molloi Sabee, Fayad Zahi, Kakadiaris Ioannis, Abela George, Narula Jagat, Williams Kim, Shah Prediman, Benjamin Emelia, Atlas Kyle, Levy Daniel, Zhang Chenyu, Reeves Anthony, Atlas Thomas, Yankelevitz David, Mirjalili Seyed Reza, Henschke Claudia

De novo lipogenesis is related to multiple cardiometabolic diseases in post-myocardial infarction patients of the Alpha Omega Cohort

Heerkens Luc, Pertiwi Kamalita, Van Duijnhoven Franzel, Geleijnse Johanna

More abstracts from these authors:
Vessel-Specific CT Calcium Scoring in Peripheral Artery Disease Using Deep Learning

Musini Kumudha, Rimmerman Eleanor, Chou Ting-heng, Shin Kyle, Bobbey Adam, Atway Said, Corriere Matthew, Go Michael, Stacy Mitchel

Dynamic PET Imaging Quantifies Absolute Measures of Calf Muscle Perfusion in a Porcine Model of Peripheral Endovascular Arterial Occlusion and PAD Patients with Claudication

Rimmerman Eleanor, Musini Kumudha, Shin Kyle, Jayaraman Anjana, Tomko Patrick, Go Michael, Stacy Mitchel

You have to be authorized to contact abstract author. Please, Login
Not Available