Title :
The contribution of artificial intelligence tools in screening for cancer of the cervix
Author :
Guesmi, L. ; Nabli, L. ; Mohamed, B.H.
Author_Institution :
Ecole Nat. Eng. of Monastir, Lab. Technol. & Med. Imaging (TIM), Monastir, Tunisia
Abstract :
Recently, research into pathological cytology were intended to put in places of artificial intelligence systems based on the development of new diagnostic technologies and the cell image segmentation. These technologies are not intended to substitute the human expert but to facilitate his task. The objective of this work is to develop a method for diagnosing cancer cervical smears using cervical - vaginal segmented to build our database and a human supervisor and as an automatic tool manage and monitor the execution of the operation of diagnostic and proposing corrective actions if necessary. The Supervisor Smart is manufactured by the technique of neural networks with a success rate of 43.3% followed by the technique of fuzzy logic with a success rate equal to 56.7% and finally to improve this rate we used neuro-fuzzy approach which has a rate which reaches 94%.
Keywords :
artificial intelligence; cancer; diagnostic expert systems; fuzzy logic; image segmentation; medical image processing; neural nets; patient diagnosis; artificial intelligence tool system; automatic tool management; cancer cervical smear diagnosing; cell image segmentation; cervix cancer; corrective actions; diagnostic technology; fuzzy logic; human expert; neural network; neuro-fuzzy approach; pathological cytology; vaginal segmentation; Cancer; Humans; Image segmentation; Predictive models; Shape; Cancer of the cervix; artificial intelligence; fuzzy logic; neural networks; segmentation of cervical smear - vaginal (CSVs); supervisor;
Conference_Titel :
Communications, Computing and Control Applications (CCCA), 2011 International Conference on
Conference_Location :
Hammamet
Print_ISBN :
978-1-4244-9795-9
DOI :
10.1109/CCCA.2011.6031230