The research group at the Central Computational and Numerical Sciences Division, Institute of Advanced Study in Science and Technology, Guwahati, led by Dr Lipi B Mahanta, has developed artificial intelligence (AI) based algorithms as an evolutionary aid to rapid diagnosis and prediction of oral squamous cell carcinoma including its grading.
The scenario of Oral cancer in India is very gloomy. With 16.1% of all cancers amongst men and 10.4% amongst women belonging to this domain, the picture is all the more alarming in the NE India where statistics reveal 44.5% for men and 16.7% for women. Further, cancer of the buccal cavity is predominant (80% cases in the study region) and 90% are diagnosed with squamous cell carcinoma. Hospitals and laboratories of the region are reeling under the burden of patients, and rapid delivery of diagnosis is a privilege only a finger count who can afford.
The advent of deep learning in AI holds an extraordinary prospect in digital image analysis to serve as a computational aid in the diagnosis of cancer, thus providing help in timely and effective prognosis and multi-modal treatment protocols for cancer patients and reducing the operational workload of pathologists while enhancing management of the disease. With this motivation, the group has been working persistently in many domains of cancer diagnosis, with close connections and collaborations with reputed cancer hospitals and pathological laboratories of the region.
Oral cavity cancers are also known to have a high recurrence rate compared to other cancers due to the high consumption of betel nut and tobacco. This cancer group is characterized by epithelial squamous tissue differentiation and aggressive tumour growth disrupting the basement membrane of the inner cheek region and thus can be graded by Broder’s histopathological system as well-differentiated SCC (WDSCC), moderately differentiated SCC (MDSCC) and poorly differentiated SCC (PDSCC). The cellular morphometry highlighting the tumour growth displays a very minute histological difference separating the three classes which are very hard to capture by the human eye. It has remained elusive due to its highly similar histological features which even pathologists find difficult to classify.
The study was a challenge due to unavailability of any benchmark oral cancer dataset for the study, and thus the scientists generated an indigenous dataset through the collaborations. Exploring different state-of-the-art AI techniques and playing with their proposed method, the scientists have gained unprecedented accuracy in oral cancer grading. The study is conducted applying two approaches: (i) through the application of transfer learning using pre-trained deep convolutional neural network (CNN) wherein four candidate pre-trained models, namely Alexnet, VGG-16, VGG-19 and Resnet-50, were chosen to find the most suitable model for our classification problem, and (ii) by a proposed CNN model
developed to fit the problem (Figure). Although the highest classification accuracy of 92.15% is achieved by Resnet-50 model, the experimental findings highlight that the proposed CNN model outperformed the transfer learning approaches displaying an accuracy of 97.5%. The work has been published in Neural Networks (Volume 128, August 2020) https://doi.org/10.1016/j.neunet.2020.05.003.
As of now, the tables are set for converting the algorithm into proper software to move on to carry out field trials. This is the next challenge that the group is prepared to meet, considering the ever-present gap between the health and IT sector. Dr Mahanta aspires for all the advanced infrastructural support to meet these challenges and feels that the software needs to be actively tested in hospitals all over the nation, to make it truly robust, more accurate and real-time worthy.
Dr Lipi B Mahanta (Mentor) – email@example.com;
Ms Tabassum Yesmin Rahman — firstname.lastname@example.org;
Ms Elima Hussain — email@example.com ;
Mr Navarun Das — firstname.lastname@example.org.