Contributed by: Abin Banerjee
Working with Deloitte Consulting India Pvt Ltd with 15 +years of Experience.
The Capstone was indeed was a great experience of the PGP AIFL program.
Our capstone project topic was ‘Creating Value for life – AI-based platform for early detection of Medical Conditions’ – further to it; we focused specifically on the below area:
‘Early Detection of Skin Cancer – Solution for Identifying and Defining Skin Cancers using AI’
We came up with the topic after brainstorming with the group members on how AI can be best utilized to create value for life. Our objective was also to leverage the knowledge we gained through the training and to apply that to a practical problem.
We found that Skin Cancer is one of the most hazardous forms and common types of cancer in the world. Each year there are approximately more than 10 million new cases of skin cancer recorded globally – this number is alarming. The survival rate is very low if diagnosed in later stages.
We realized that Artificial Intelligence could play a very important role in using Medical Image Diagnosis to detect this disease in early stages. However, the AI systems for the classification of different skin lesions are still in the early stages of clinical application in terms of being ready to aid in diagnosing skin cancers. Moreover, not many players are doing research in this direction for conditions specific to the Indian subcontinent.
Hence, we chose this topic to explore the potential of applying AI in this space, and thus this topic was arrived at.
In our solution named ‘HAIS’ that we have proposed through our Capstone project, we’ve planned to apply Deep CNN techniques, a proven technique in the image classification space. We took cognizance that applying high-capacity Deep CNN in suspicious pigmented lesions (SPLs) analysis had always been a challenge because of the scarcity of labeled data. So, to overcome it, our HAIS solution is proposed to include two primary methods:
First, a classification model to improve the performance of classification of skin lesion using Deep CNN and Data Augmentation.
Second, use of image data augmentation for overcoming the problem of data limitation and examine the influence of the different numbers of augmented samples on the performance of different classifiers.
Through this topic, we explored the advancements in the digital image-based AI solutions for skin cancer diagnosis, along with some challenges and future opportunities to improve these AI systems to support dermatologists and enhance their ability to diagnose skin cancer in early stages.
It was great learning not only from the functional aspects of it but, more importantly, how AI can be leveraged to come up with a solution to a real-life challenge and how it can create value for life.
The Capstone was an experience to treasure for life – we as a team thoroughly enjoyed this learning experience. Not only did we learn about the topic and apply our learning from training to design SI based solutions for it, but we also made some friends for life – a great network to fall back on in the future – Overall, it was a great GREAT experience – thankful to Great leaning for providing this opportunity and the great learning experience.
The overall PGP AIFL program learning journey was very enriching and enlightening. I felt the course was designed very thoughtfully, and the materials are very relevant for the roles we play in the corporate world. Needless to say, we all benefited from a great set of faculty members and industry experts – overall – a great journey thus far.
I think the experience from the capstone project and overall learning journey has made me better equipped to apply this at my workplace – I’ve thoughts and plans to take it forward and apply it in analyzing and designing solutions in the future.
From the bottom of my heart, I want to thank each and every faculty member, industry expert, and coordinator for making this a memorable journey and a great learning experience.
I would also take this opportunity to recognize our program coordinator, Ms. Priya Bohra, for being so accommodating at the same time being truly professional in her role.
Last but not least, thanks to each of my team members for making this a memorable learning experience – we made friends for life.
I would surely share my experience with my friends and encourage them to take this learning experience.
