Prevention of diabetes in rural India with a telemedicine intervention

Background : Diabetes care is not presently available, accessible, or affordable to people living in rural areas in developing countries, such as India. The Chunampet Rural Diabetes Prevention Project (CRDPP) was conceived with the aim of implementing comprehensive diabetes screening, prevention, and treatment using a combination of telemedicine and personalized care in rural India.

Methods : This project was undertaken in a cluster of 42 villages in and around the Chunampet village in the state of Tamil Nadu in southern India. A telemedicine van was used to screen for diabetes and its complications using retinal photography, Doppler imaging, biothesiometry, and electrocardiography using standardized techniques. A rural diabetes center was set up to provide basic diabetes care.

Results : Of the total 27,014 adult population living in 42 villages, 23,380 (86.5%) were screened for diabetes, of which 1138 (4.9%) had diabetes and 3410 (14.6%) had prediabetes. A total of 1001 diabetes subjects were screened for complications (response rate of 88.0%). Diabetic retinopathy was detected in 18.2%, neuropathy in 30.9%, microalbuminuria in 24.3%, peripheral vascular disease in 7.3%, and coronary artery disease in 10.8%. The mean hemoglobin A1c levels among the diabetes subjects in the whole community decreased from9.3 ± 2.6% to 8.5 ± 2.4% within 1 year. Less than 5% of patients needed referral for further management to the tertiary diabetes hospital in Chennai.

Multichannel biomedical time series clustering via hierarchical probabilistic latent semantic analysis

Biomedical time series clustering that automatically groups a collection of time series according to their internal similarity is of importance for medical record management and inspection such as bio-signals archiving and retrieval. In this paper, a novel framework that automatically groups a set of unlabelled multichannel biomedical time series according to their internal structural similarity is proposed. Specifically, we treat a multichannel biomedical time series as a document and extract local segments from the time series as words. We extend a topic model, i.e., the Hierarchical probabilistic Latent Semantic Analysis (H-pLSA), which was originally developed for visual motion analysis to cluster a set of unlabelled multichannel time series. The H-pLSA models each channel of the multichannel time series using a local pLSA in the first layer. The topics learned in the local pLSA are then fed to a global pLSA in the second layer to discover the categories of multichannel time series. Experiments on a dataset extracted from multichannel Electrocardiography (ECG) signals demonstrate that the proposed method performs better than previous state-of-the-art approaches and is relatively robust to the variations of parameters including length of local segments and dictionary size. Although the experimental evaluation used the multichannel ECG signals in a biometric scenario, the proposed algorithm is a universal framework for multichannel biomedical time series clustering according to their structural similarity, which has many applications in biomedical time series management.