TY - GEN
T1 - Design of a sign language transformer to enable the participation of persons with disabilities in remote healthcare systems for ensuring universal healthcare coverage
AU - Podder, Kanchon Kanti
AU - Tabassum, Shaira
AU - Khan, Ludmila Emdad
AU - Salam, Khan Md Anwarus
AU - Maruf, Rafiqul Islam
AU - Ahmed, Ashir
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021/5/17
Y1 - 2021/5/17
N2 - Poverty, Rurality and Disability are the three major burdens in achieving Universal Healthcare Coverage (UHC). The advent of Information and Communication Technologies (ICT) and remote healthcare systems play a significant role to reach the unreached communities and are addressing rurality and poverty issues. However, the persons with disability (PWDs), especially the speech and hearing impaired people find it difficult to participate in remote healthcare systems as they cannot communicate with a remote doctor. A design of a 'Sign Language Transformer (SLT)' has been introduced in this paper for the patients who know sign languages to establish a communication with a remote doctor who cannot interpret such signs. The primary function of this SLT is to recognize the signs/gestures from video images and translate them into both text and speech (SLTT), and to translate doctor's speech into sign language (STSL). Sign representation of words and sentences requires hand gesture, movement, and orientation. Several technologies such as the two-stream CNN, the two-stream 3D CNN, the LSTM, the 3DCNN+ ConvLSTM, the 3D CNN and the 3D CNN + LSTM are commonly used techniques to recognize human gestures. The proposed SLT model will evaluate the performances of these technologies to transform the Bangla Sign Language and to recommend the suitable technology for designing a sign language transformer.
AB - Poverty, Rurality and Disability are the three major burdens in achieving Universal Healthcare Coverage (UHC). The advent of Information and Communication Technologies (ICT) and remote healthcare systems play a significant role to reach the unreached communities and are addressing rurality and poverty issues. However, the persons with disability (PWDs), especially the speech and hearing impaired people find it difficult to participate in remote healthcare systems as they cannot communicate with a remote doctor. A design of a 'Sign Language Transformer (SLT)' has been introduced in this paper for the patients who know sign languages to establish a communication with a remote doctor who cannot interpret such signs. The primary function of this SLT is to recognize the signs/gestures from video images and translate them into both text and speech (SLTT), and to translate doctor's speech into sign language (STSL). Sign representation of words and sentences requires hand gesture, movement, and orientation. Several technologies such as the two-stream CNN, the two-stream 3D CNN, the LSTM, the 3DCNN+ ConvLSTM, the 3D CNN and the 3D CNN + LSTM are commonly used techniques to recognize human gestures. The proposed SLT model will evaluate the performances of these technologies to transform the Bangla Sign Language and to recommend the suitable technology for designing a sign language transformer.
UR - http://www.scopus.com/inward/record.url?scp=85112178989&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85112178989&partnerID=8YFLogxK
U2 - 10.1109/TEMSCON-EUR52034.2021.9488605
DO - 10.1109/TEMSCON-EUR52034.2021.9488605
M3 - Conference contribution
AN - SCOPUS:85112178989
T3 - 2021 IEEE Technology and Engineering Management Conference - Europe, TEMSCON-EUR 2021
BT - 2021 IEEE Technology and Engineering Management Conference - Europe, TEMSCON-EUR 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 IEEE Technology and Engineering Management Conference - Europe, TEMSCON-EUR 2021
Y2 - 17 May 2021 through 20 May 2021
ER -