Yazar "Aydin, Muhammed Ali" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim
    Öğe
    Identification of phantom movements with an ensemble learning approach
    (Elsevier Ltd, 2022) Akbulut, Akhan; Gungor, Feray; Tarakci, Ela; Aydin, Muhammed Ali; Zaim, Abdul Halim; Catal, Cagatay
    Phantom limb pain after amputation is a debilitating condition that negatively affects activities of daily life and the quality of life of amputees. Most amputees are able to control the movement of the missing limb, which is called the phantom limb movement. Recognition of these movements is crucial for both technology-based amputee rehabilitation and prosthetic control. The aim of the current study is to classify and recognize the phantom movements in four different amputation levels of the upper and lower extremities. In the current study, we utilized ensemble learning algorithms for the recognition and classification of phantom movements of the different amputation levels of the upper and lower extremity. In this context, sEMG signals obtained from 38 amputees and 25 healthy individuals were collected and the dataset was created. Studies of processing sEMG signals in amputees are rather limited, and studies are generally on the classification of upper extremity and hand movements. Our study demonstrated that the ensemble learning-based models resulted in higher accuracy in the detection of phantom movements. The ensemble learning-based approaches outperformed the SVM, Decision tree, and kNN methods. The accuracy of the movement pattern recognition in healthy people was up to 96.33%, this was at most 79.16% in amputees.
  • Küçük Resim Yok
    Öğe
    MOBILE VISUAL ACUITY ASSESSMENT APPLICATION: AcuMob
    (Istanbul Univ, Fac Engineering, 2017) Akbulut, Akhan; Aydin, Muhammed Ali; Zaim, Abdul Halim
    This paper presents a mobile healthcare (mHealth) system for estimation of visual impairment that provides easiness by specifying the degree of an eye as orthoscopes. Our proposed system called AcuMob which is an Android based mobile application aimed to be used by patients who have myopia. In the crowd society, our proposed app will be implemented faster than the traditional ophthalmologic examination treatments as an alternative. Because AcuMob can be used in everywhere in any time slot, it is offered in the area where the ophthalmologist is not available. The system is developed with using Xamarin framework and voice commands are used to interact with mobile app. Some preferable letters that are suggested by the ophthalmologists were used in the system. The letter categories are specified according to letters' sizes. In the start-up screen, the biggest letter is demonstrated and if the user responds correct answer, the letter's size is being smaller. However, if the user says wrong answer three times consecutively, eyesight ratio is produced by the system to the user referencing to Snellen Chart's information. This article has aimed at making a prediction about the visual impairment's degree. Thanks to AcuMob, people can get idea about their visual acuity without consulting to an eye medical doctor (MD). For the evaluation of systems' reliability, field tests were performed at Bayrampasa Goz Vakfi Hospital in Istanbul with two ophthalmologist specialists. At the end of trials, the actual diagnosed degrees and the equivalent degree of eyesight ratios according to Snellen Chart's information is compared and the success rates are shown. The system achieved at the 65% of average success rate, which can give users an idea about current condition of their visions.