Vision-based Therapeutic System Involving Finger Exercise

Azad, Nafis Saami; Muhaimin, Al; Zoyee, Maliha Mehzabin

dc.contributor.author	Azad, Nafis Saami
dc.contributor.author	Muhaimin, Al
dc.contributor.author	Zoyee, Maliha Mehzabin
dc.date.accessioned	2023-03-24T08:26:17Z
dc.date.available	2023-03-24T08:26:17Z
dc.date.issued	2022-05-30
dc.identifier.citation	[1] Easy Hand Exercises to Boost Recovery from a Stroke shorturl.at/npG79 [2] A Gentle Introduction to Ensemble Learning Algorithms https://machinelearningmastery.com/tour-of-ensemble-learning-algorithms/ [3] Bakar, M. Zabri Abu, et al. ”Computer vision-based hand deviation exercise for rehabilitation.” 2015 IEEE International Conference on Control System, Computing and Engineering (ICCSCE). IEEE, 2015. [4] Decker, Jilyan, et al. ”Wiihabilitation: rehabilitation of wrist flexion and extension using a wiimote-based game system.” Governor’s School of Engineering and Technology Research Journal (2009): 92-98. [5] Strong, Kathleen, Colin Mathers, and Ruth Bonita. ”Preventing stroke: saving lives around the world.” The Lancet Neurology 6.2 (2007): 182-187. [6] Langhorne, Peter, Julie Bernhardt, and Gert Kwakkel. ”Stroke rehabilitation.” The Lancet 377.9778 (2011): 1693-1702. [7] Carpinella, Ilaria, Johanna Jonsdottir, and Maurizio Ferrarin. ”Multi-finger coordination in healthy subjects and stroke patients: a mathematical modelling approach.” Journal of neuroengineering and rehabilitation 8.1 (2011): 1-20. [8] Pompili, Giulia, et al. ”Development of a Low-cost Glove for Thumb Rehabilitation: Design and Evaluation.” 2020 IEEE International Conference on Human-Machine Systems (ICHMS). IEEE, 2020. [9] Nuzzi, Cristina, et al. ”HANDS: an RGB-D dataset of static hand-gestures for human-robot interaction.” Data in Brief 35 (2021): 106791 [10] Kaczmarek, Piotr, Tomasz Ma´nkowski, and Jakub Tomczy´nski. ”putEMG—a surface electromyography hand gesture recognition dataset.” Sensors 19.16 (2019): 3548. 37 [11] Hand Gesture Recognition Database https://www.kaggle.com/datasets/gti-upm/leapgestrecog [12] Qurratu’aini, Dayang, et al. ”Visual-based fingertip detection for hand rehabilitation.” Indonesian Journal of Electrical Engineering and Computer Science 9.2 (2018): 474-480. [13] Zhang, Yin, Rong Jin, and Zhi-Hua Zhou. ”Understanding bag-of-words model: a statistical framework.” International Journal of Machine Learning and Cybernetics 1.1 (2010): 43-52. [14] Lei, Xinyu, Hongguang Pan, and Xiangdong Huang. ”A dilated CNN model for image classification.” IEEE Access 7 (2019): 124087-124095. [15] Agarap, Abien Fred. ”An architecture combining convolutional neural network (CNN) and support vector machine (SVM) for image classification.” arXiv preprint arXiv:1712.03541 (2017). [16] Helpful Hand Exercises for Stroke Patients of All Ability Levels https://www.flintrehab.com/hand-exercises-for-stroke-patients/ [17] Stroke (Cerebral Vascular Accident (CVA) and Spinal Stroke) https://www.christopherreeve.org/living-with-paralysis/health/ causes-of-paralysis/stroke	en_US
dc.identifier.uri	http://hdl.handle.net/123456789/1784
dc.description	Supervised by Mr. Mohammad Ridwan Kabir, Lecturer, Department of Computer Science and Engineering(CSE), Islamic University of Technology (IUT) Board Bazar, Gazipur-1704, Bangladesh. This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Software Engineering, 2022.	en_US
dc.description.abstract	This work aims to create a system that will allow patients to receive physiotherapy at home. We focus on two key parts, a vision based approach, and rehabilitation therapy primarily for stroke patients with upper limb disability through finger exercises. Our hardware consists of only using a HD webcam or a camera. For our hand detection, we are using three deep learning models. The first model is trained using Google’s Mediapipe hands to find the hand landmark from an image, and using the landmarks to recognize the therapeutic hand gestures. To train our model, we also collected an extensive dataset for the pinching finger exercises with different environmental conditions, distances from camera and orientations. The data we collected have been used to create 3 deep learning models capable of detecting therapeutic hand gestures from real-time video feeds. In this report, we have discussed the methodologies we used for data collection, the types of models we built and their architecture along with their strong points and shortcomings. Lastly, we elaborated on the future works that we intend to perform.	en_US
dc.language.iso	en	en_US
dc.publisher	Department of Computer Science and Engineering(CSE), Islamic University of Technology(IUT), Board Bazar, Gazipur-1704, Bangladesh	en_US
dc.subject	Computer Vision; Therapeutic Finger Exercises; Deep-learning; Convolutional Neural Network; Rehabilitation; Stroke Patient Therapy	en_US
dc.title	Vision-based Therapeutic System Involving Finger Exercise	en_US
dc.type	Thesis	en_US