TY - CHAP
T1 - Post COVID-19 Remote Medicine and Telemedicine Evaluation via Natural Language Processing Techniques
AU - Trovati, Marcello
AU - Soussan, Tariq
AU - Korkontzelos, Yannis
AU - Polatidis, Nikolaos
PY - 2024/5/13
Y1 - 2024/5/13
N2 - The COVID-19 pandemic greatly increased the workload of hospitals and various restrictions were put in place to restrict the spread of the virus. As a result, clinics were shut down and patient interaction with the clinics and hospitals shifted to different forms of applications of telemedicine. Although telemedicine was in use in different parts of the world before the pandemic, it was mainly being used by people who were in remote locations and unable to access the healthcare facilities due to travel or certain medical disability. This also gave rise to opportunities to carry out improvement in telemedicine platforms and the ways by which healthcare professionals can measure and deliver more patient-centric healthcare services. One way is to use text mining techniques to extract the right words and their corresponding relations upon which a deep learning model healthcare application can be built to aid in the decision making of clinicians. As the patient data collected by the healthcare professional is mostly in the form of free text and the medical language being highly specialised, finding the relevant concepts and combination of words is quite challenging. In this chapter, we have extracted combination of words from PubMed database and tweets from Twitter related to telemedicine and remote consultations by using text mining techniques and Improved Sentiment Urgent Emotion Detection (ISUED) Model from a previous experiment to find the right combination of words. By using such techniques, we were able to determine the right semantic relation between the specified words to help detect the polarity from the unstructured data.
AB - The COVID-19 pandemic greatly increased the workload of hospitals and various restrictions were put in place to restrict the spread of the virus. As a result, clinics were shut down and patient interaction with the clinics and hospitals shifted to different forms of applications of telemedicine. Although telemedicine was in use in different parts of the world before the pandemic, it was mainly being used by people who were in remote locations and unable to access the healthcare facilities due to travel or certain medical disability. This also gave rise to opportunities to carry out improvement in telemedicine platforms and the ways by which healthcare professionals can measure and deliver more patient-centric healthcare services. One way is to use text mining techniques to extract the right words and their corresponding relations upon which a deep learning model healthcare application can be built to aid in the decision making of clinicians. As the patient data collected by the healthcare professional is mostly in the form of free text and the medical language being highly specialised, finding the relevant concepts and combination of words is quite challenging. In this chapter, we have extracted combination of words from PubMed database and tweets from Twitter related to telemedicine and remote consultations by using text mining techniques and Improved Sentiment Urgent Emotion Detection (ISUED) Model from a previous experiment to find the right combination of words. By using such techniques, we were able to determine the right semantic relation between the specified words to help detect the polarity from the unstructured data.
U2 - 10.1007/978-3-031-56818-3_1
DO - 10.1007/978-3-031-56818-3_1
M3 - Chapter
SN - 9783031568176
SN - 9783031568206
T3 - Signals and Communication Technology
SP - 3
EP - 21
BT - Data Science and Artificial Intelligence for Digital Healthcare
A2 - Singh, Pradeep Kumar
A2 - Trovati, Marcello
A2 - Murtagh, Fionn
A2 - Atiquzzaman, Mohammed
A2 - Farid, Mohsen
PB - Springer
CY - Switzerland
ER -