EXPLAINABLE AI FOR COLON POLYPS DETECTION
Artificial intelligence (AI) has rapidly developed over the past few decades and was considered to have a huge effect on all forms of technologies and everyday life. Due to the large volume of data, the usage of AI in healthcare system, has been increasing rapidly. Different AI techniques have be...
| Main Author: | Ramli, Nur Iman Athilah |
|---|---|
| Format: | Final Year Project |
| Language: | English |
| Institution: | Universiti Teknologi Petronas |
| Record Id / ISBN-0: | utp-utpedia.23033 / |
| Published: |
Universiti Teknologi PETRONAS
2021
|
| Subjects: | |
| Online Access: |
http://utpedia.utp.edu.my/23033/1/EE102_24600_Nur%20Iman%20Athilah%20Binti%20Ramli.pdf http://utpedia.utp.edu.my/23033/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: |
Artificial intelligence (AI) has rapidly developed over the past few decades and was
considered to have a huge effect on all forms of technologies and everyday life. Due to the
large volume of data, the usage of AI in healthcare system, has been increasing rapidly.
Different AI techniques have been used in medical imaging, including, machine learning and
the deep learning technique which is convolutional neural network (CNN) which have helped
doctors identify diseases accurately and assess effective care for them. The use and processing
of a large number of the digital images and a lot of medical records over a period of time has
contributed to the development of big data. The interest in applying computer-aided diagnosis
(CAD) and artificial intelligence (AI) to endoscopic evaluation in the gastrointestinal tract has
been revived by recent developments in computing power, coupled with rapid growth in the
quantity and availability of data. However, regulatory impediments that need to be implement
before applying CAD in routine clinical practise because the complexity of AI driven clinical
decision-making, which is generally regarded as a ' black box.' And as such, this study aims to
create an explainable AI-based CAD method to help endoscopists distinguish polyps during
endoscopy, which can justify their decisions using classification rules. |
|---|