You are here
Helping doctors make faster and better decisions
ALTHOUGH medical imaging contributes over 90 per cent of the data used to diagnose patient conditions, there is a global shortage of healthcare professionals such as radiologists and imaging specialists to analyse this information. This results in long waiting times for patients, poor access to high-quality diagnoses and long working hours for medical professionals. Adding to the problem are cases of misdiagnoses due to human error that lead to less than optimal treatments.
To address this problem, BioMind, an artificial intelligence (AI) company specialising in healthcare, developed an application that analyses medical images such as magnetic resonance imaging (MRI) and computed tomography (CT) images and reports the analysis automatically.
The company employs a team of over 60 deep learning scientists, medical experts, and research advisers.
The application covers brain-related conditions including stroke, vascular diseases and tumours and stroke-related medical conditions, which make up the majority of the conditions determined through medical imaging.
Through the use of AI, BioMind is able to identify abnormalities on medical image scans, recommend possible diagnoses and generate a report for doctors to review, all within seconds. According to BioMind CEO Raymond Moh, this is faster than the average one hour or more that doctors usually take for analysis and reporting.
What's more, BioMind has the ability to self-learn and improve through various cycles of predictions and doctors' revisions. This helps doctors to reduce misdiagnoses, speeds up the process of diagnosis, and reduces waiting times for patients.
"At present, there are no existing solutions that can deal with these issues effectively. Machine learning, and in particular, deep learning, is a powerful technology that can be harnessed to personalise healthcare solutions and deal with the high volume of routine analysis," said Mr Moh. BioMind was named the winner of the Most Promising Innovation (Gold) award at this year's SG:D Techblazer Awards.
"By continuously learning from past data and the best patient outcomes, our machine can assist doctors to make better and faster decisions. By scaling this 'machine intelligence' and deploying it globally, we help patients to access high quality medical services without them having to travel to good hospitals in big cities."
The development of BioMind began in 2017, and was officially launched in June 2018 when it obtained the European CE mark certification for clinical use. Soon after, the application won the CHAIN Cup, billed as the world's first competition in neuroimaging between AI and human experts. In the competition, BioMind went toe-to-toe with real doctors to diagnose brain tumours and predict the expansion of brain haematomas or bruises.
The company has also received approval from Singapore's Health Sciences Authority (HSA) for the use of its application.
Mr Moh noted that the biggest challenge now is to combine both the machine learning capabilities of BioMind with the clinical expertise of medical professionals.
"Deep understanding and integration of clinical knowledge is necessary to co-create the product so that it can be suitable for clinical use. We also need to work very closely with the stakeholders including hospital users and regulators who are new to the deep and powerful technology," he said.
Looking ahead, BioMind will be expanded to analyse medical image scans for conditions in all parts of the body, effectively becoming a "multi-specialist" assistant for doctors.
CASHSHIELD: MOST PROMISING INNOVATION (Silver)
TO help companies deal with fraud, enterprise risk management company CashShield has developed a solution that leverages artificial intelligence to automatically detect attacks such as loan fraud or money laundering in real time. The system requires no human intervention and covers all vulnerable entries to fraud attacks across multiple channels and devices. To do so, it profiles more than 5 billion devices and over 500 million user accounts every year.
CashShield helps to boost the security of businesses from around the world, from e-wallet and super app providers to e-commerce, travel and digital goods players. The company's clients include global enterprises such as OVO, Alibaba, Razer, Scalefast and Grab.
"By always being ahead of fraud, CashShield gives enterprises a higher level of confidence to scale and grow their revenue potential, maximising their business growth," said CashShield chief technology officer Khin Kai Loke.
Going forward, the company aims to consolidate its position as a leader in providing cyber fraud solutions in Asia, expand its footprint in the US and Europe, as well as penetrate new markets in India and Latin America.
TOOKITAKI HOLDINGS: MOST PROMISING INNOVATION (Silver)
Software firm Tookitaki is a specialist in creating machine learning-powered solutions to maximise efficiency and reduce risks in the compliance process for the financial services sector. One of its offerings is an Anti-Money Laundering Suite (AMLS) that is designed to mitigate the risk of money laundering and reduce the cost of compliance.
AMLS comprises two modules; one for transaction monitoring and another for names and sanctions screening. The application improves risk management by identifying suspicious transactions and activity and alerting users to them. It also provides detailed explanations for every alert to ensure better regulatory compliance. According to Tookitaki, the solution is also able to integrate seamlessly with existing anti-money laundering systems.
SPARK SYSTEMS: MOST PROMISING INNOVATION (Bronze)
Fintech company Spark Systems provides solutions for the forex (FX) trading community. The company's SparkFX trading platform connects buyers and sellers directly, and aims to improve efficient access to FX liquidity and reduce trading costs.