Bangalore-based startup Atom360 is the youngest on the block in the AI space and the only Indian entrant that progressed to second round of the famed IBM Watson X prize that kick-started in 2016. The $5 million IBM Watson AI XPrize competition, which kicked off in 2016 and will end in 2020, is the interestingly the first of the X-prize contests to feature a contestant-defined “open” goal rather than a predetermined objective.
Atom360 made it to the second round of the competition and was one of the 59 teams, representing 14 countries. While the top 10 teams were recognized at the Annual Conference on Neural Information Processing Systems in December 2017 – Atom360 that submitted an entry related to the healthcare domain also received a superb response.
Talking to Analytics India Magazine about the AI-focused healthcare solution, founder and CEO Reuben Fernandes said, “We feel great to be representing our country in this global AI competition. As part of the competition, we are working on AI-assisted surgical robot. We have broken this problem down and are trying to get market validation for each algorithm. The application ranges from tumor segmentation in medical data to reducing oral cancer deaths in India using deep learning algorithms”.
The X-Prize challenge led to the inception of this startup
This Bangalore-based six-member startup is currently working in a stealth mode and would be launching two products this year. The startup was incepted at the time when IBM Watson X Prize Challenge was announced, after their healthcare entry was selected, the team started working full-time in 2017. Atom360 has a very young team, the founder, Fernandes is a 2012 SRM University, Chennai graduate who had always been interested in robotics and deepened his Deep Learning base with online programs. Meanwhile, the average age of the team is around 24 years with everyone interested in making an impact in the world using AI, revealed Fernandes.
Atom360’s AI solution can reduce deaths occurring due to oral cancer
The impact of AI on healthcare has been huge – a 2016 Frost & Sullivan research forecasted that the market for AI in healthcare is projected to reach $6.6 billion by 2021. From healthcare bots to virtual assistants and AI-powered diagnostic tools, AI has the potential to improve medical outcome by 30-40 percent and reduce cost of treatment by 50 per cent. When AIM quizzed Fernandes on focusing on the healthcare domain, he said, “We believe in human and AI working together to improve healthcare. We work on augmented intelligence making solutions for doctors therefore improving efficiency and reducing time.”
The startup aims to provide a AI-powered solution for oral cancer, one of the top three types of cancers in India. A recent study indicated that India also has among the highest and most number of avoidable deaths due to cancer. In India, tobacco-related diseases cost an astounding INR 100 billion to India’s healthcare, a news report indicates. “The impact we want to create is reduce the deaths due to oral cancer by at-least half before 2020. Getting our doctors to be augmented by AI will help bring quality care to all Indians at low costs,” shared Fernandes.
For Atom360, the application ranges from tumor segmentation in medical data to reducing Oral Cancer deaths in India using deep learning algorithms. Also, the final solution is the surgical robot. “We are selectively getting into domains such as robotics, tumor segmentation and computer vision so that can validate our algorithms that we will eventually use in our surgical robot,” he shared.
The startup is working with algorithms ranging from reinforcement learning to deep convolution neural net. “We use Deep learning for Image classification and segmentation to detecting tumors in medical images,” he added. Are the solutions scalable? “Well, part of the judgement criteria during the X Prize challenge was measuring the impact we will be making in the world. Our solutions are highly scalable and we will be launching two of our products the beginning of the next quarter,” he shared.
Implementing the AI Solution
The solution is targeted at radiologists who can use to segment tumour faster and more efficiently. Besides, Fernandes added that the app would enable any person or volunteer to detect oral cancer through the app for periodic self-diagnosis. When quizzed about time to market, he said, “We will be deploying systems for cancer segmentation and oral cancer screening this year. Our clients are Doctors we are starting with radiologists as our clients who can leverage the sensing capability of the surgical robot,” he said.
The startup’s augmented AI solution would help doctors screen and segment the tumour effectively and also reduce the time take up by the segmentation procedure significantly.
How The X Prize Challenge Gave Global Visibility
The competition that led to the founding of the startup gave the young team global visibility and also globally connecting them with international mentors and talented people from India who want to be a part of their exciting journey. We also have support in terms of cloud computing credits to help accelerate fast, shared Fernandes. When prodded about his hiring criteria, he shared, “We hire young talented people who we think have the potential to learn fast. We have leaders from the AI field as our mentors who guide us to accomplish our goals”.
The founder’s advice to Deep Learning enthusiasts who want to shift towards this field is to learn from the courses available on the Internet, there are new ones coming every month and one can go through courses and research papers to get a different perspective.
Try deep learning using MATLAB