학술논문
Optimizing Rare Disease Gait Classification through Data Balancing and Generative AI: Insights from Hereditary Cerebellar Ataxia.
Document Type
Academic Journal
Author
Trabassi D; Department of Medical and Surgical Sciences and Biotechnologies, 'Sapienza' University of Rome, 04100 Latina, Italy.; Castiglia SF; Department of Medical and Surgical Sciences and Biotechnologies, 'Sapienza' University of Rome, 04100 Latina, Italy.; Department of Brain and Behavioral Sciences, University of Pavia, 27100 Pavia, Italy.; Bini F; Department of Mechanical and Aerospace Engineering, Sapienza University of Rome, 00184 Rome, Italy.; Marinozzi F; Department of Mechanical and Aerospace Engineering, Sapienza University of Rome, 00184 Rome, Italy.; Ajoudani A; Department of Advanced Robotics, Italian Institute of Technology, 16163 Genoa, Italy.; Lorenzini M; Department of Advanced Robotics, Italian Institute of Technology, 16163 Genoa, Italy.; Chini G; Department of Occupational and Environmental Medicine, Epidemiology and Hygiene, INAIL, Monte Porzio Catone, 00078 Rome, Italy.; Varrecchia T; Department of Occupational and Environmental Medicine, Epidemiology and Hygiene, INAIL, Monte Porzio Catone, 00078 Rome, Italy.; Ranavolo A; Department of Occupational and Environmental Medicine, Epidemiology and Hygiene, INAIL, Monte Porzio Catone, 00078 Rome, Italy.; De Icco R; Department of Brain and Behavioral Sciences, University of Pavia, 27100 Pavia, Italy.; Headache Science & Neurorehabilitation Unit, IRCCS Mondino Foundation, 27100 Pavia, Italy.; Casali C; Department of Medical and Surgical Sciences and Biotechnologies, 'Sapienza' University of Rome, 04100 Latina, Italy.; Serrao M; Department of Medical and Surgical Sciences and Biotechnologies, 'Sapienza' University of Rome, 04100 Latina, Italy.; Movement Analysis Laboratory, Policlinico Italia, 00162 Rome, Italy.
Source
Publisher: MDPI Country of Publication: Switzerland NLM ID: 101204366 Publication Model: Electronic Cited Medium: Internet ISSN: 1424-8220 (Electronic) Linking ISSN: 14248220 NLM ISO Abbreviation: Sensors (Basel) Subsets: MEDLINE
Subject
Language
English
Abstract
The interpretability of gait analysis studies in people with rare diseases, such as those with primary hereditary cerebellar ataxia (pwCA), is frequently limited by the small sample sizes and unbalanced datasets. The purpose of this study was to assess the effectiveness of data balancing and generative artificial intelligence (AI) algorithms in generating synthetic data reflecting the actual gait abnormalities of pwCA. Gait data of 30 pwCA (age: 51.6 ± 12.2 years; 13 females, 17 males) and 100 healthy subjects (age: 57.1 ± 10.4; 60 females, 40 males) were collected at the lumbar level with an inertial measurement unit. Subsampling, oversampling, synthetic minority oversampling, generative adversarial networks, and conditional tabular generative adversarial networks (ctGAN) were applied to generate datasets to be input to a random forest classifier. Consistency and explainability metrics were also calculated to assess the coherence of the generated dataset with known gait abnormalities of pwCA. ctGAN significantly improved the classification performance compared with the original dataset and traditional data augmentation methods. ctGAN are effective methods for balancing tabular datasets from populations with rare diseases, owing to their ability to improve diagnostic models with consistent explainability.