TALK – An AAC Device: Converting Breath into Speech for the Disabled

Kunj Siddharth Dedhia

Kunj Siddharth Dedhia
Dhirubhai Ambani International School, Mumbai, Maharashtra
Title: Smartphone application based on user feedback for cyclists to reduce incidence of lower back pain
Incessant backaches are an intrinsic part of cycling. Preliminary research revealed that 45% of the cyclists in the world suffer from back pains of which 20% never recover. Although every cyclist tries to cure the pain with the help of certain stretches and exercises, prevention is always better. I propose a smartphone application based on user feedback which will alert the rider when he moves out of the correct posture for minimizing injury. Initially, I collected and analysed the rotation sensor pitch values of the lower back (ranging from – 180° to 180°) obtained from male and female, elite and trained cyclists as well as coaches from different age groups having varying heights and weight, suffering from practically no back pain. It was identified that every individual has a personalized spine posture threshold according to their comfort zones. Patterns in these threshold values were noticed among a range of age groups with a specific height and weight. Using Q Learning, the application generates an initial threshold range. During the ride, the application continuously analyses the sensor values and decreases the range over time. In addition, the rider is given an option to dismiss the alert from the application indicating he is still in his comfort zone, and consequently the feedback is incorporated. To test this, the application was used at the Gait and Movement Analysis Centre along with the 3D motion capture technology to verify improvements in body and muscle movements, especially back posture while riding.