It consisted of 30 participants diagnosed as having diabetic peripheral neuropathy. Each received traditional balance exercise, resistance exercise, and whole-body vibration, for 6 weeks.

It consisted of 30 participants diagnosed as having diabetic peripheral neuropathy. Each received traditional balance exercise, resistance exercise, and kineso tape, for 6 weeks.

All patients in both group received a traditional balance exercise program, 3 sessions per week, for 6 weeks. Each session of exercise comprised 10 min of warm-up, 40min of balance exercise, and 5min of cool down. Warm-up included treadmill walking. Balance exercise comprised two sets of sit to stand, one leg stance, tandem stance and 30 squats. The first set of exercises was performed on a stable surface, whereas the second set of exercises was performed on an unstable surface by using Thera band stability trainer. Each set of each exercise was performed for 3 min, with 1-2 min of rest in between the exercises. Cool-down included deep breathing, abdominal breathing, and mild stretching.

It was measured for all participants in both groups before and after treatment, using Neuropack S1 MEB-9004 NIHON KODEN, JAPAN nerve conduction velocity device.

It was measured for all participants in both groups before and after treatment, using Neuropack S1 MEB-9004 NIHON KODEN, JAPAN nerve conduction velocity device.

It was used to measure the balance for all participants in both groups before and after treatment. It is a 14-item list with each item consisting of a five-point ordinal scale ranging from 0 to 4, with 0 indicating the lowest level of function and 4 the highest level of function. The scale takes approximately 10 to 20 minutes to complete requiring minimal equipment (chair, stopwatch, ruler, and step) and minimal space Berg balance have high validity and reliability. Interpretation: 41-56 = low fall risk, 21-40 = medium fall risk, 0 -20 = high fall risk.

It was used to assess postural balance for all participants in both groups before and after treatment. Each subject was instructed to maintain balance on the nondominant leg with eyes opened, eyes closed, and arms spread for as long as possible. The time taken for the contralateral foot to touch the ground was measured in seconds using a stopwatch. This test was performed twice, and the highest score was recorded.

It was used to assess balance and falls risk for all participants in both groups before and after treatment. The app is freely available on the Google Play store, with participants consenting to use of their anonymized data for research and product improvement. The app uses a questionnaire on clinical falls risk factors, combined with a standing balance test and machine learning algorithms to assess balance and falls risk. Participants are provided with advice and exercises to stay healthy and help prevent falls, based on their level of risk (as determined by the smartphone based machine learning algorithm). All data were anonymized at source and transmitted via encrypted channel to a web server where they were stored (in line with GDPR and HIPAA data privacy regulations) for subsequent offline analysis. Data were extracted from the database using MySQL workbench 8.0 (Oracle, CA, USA) and analyzed using Matlab v 9.3.0 (Mathworks, Natick, VA, USA).