Radmol AI Systems
Nominated Award: Best Application of AI to achieve Social Good
Website of Company (or Linkedin profile of Person): www.radmol.com
Radmol AI is on a mission to make quality healthcare accessible to everyone by minimizing the risk of delay and error in medical diagnosis. Our solutions are driven by emerging technologies like AI ( i.e. natural language processing (NLP) and computer vision).
It was our experience during one of our pilot studies in an emerging country that truly shaped our thinking about using AI to alter the landscape of healthcare in cancer diagnosis beyond what currently exists.
During the pilot study of our Radssure solution in Rwanda, we realised that a pediatric patient who had been on the waiting list for radiology diagnosis for several months was diagnosed with cancer by one of our network of radiologists. It was sad because there was no way of know this young patient was a high-risk patient, particularly because the radiology examination he underwent was for a totally different ailment. However, this situation is not uncommon.
According to the world health organization, 80% of countries globally do not have early detection programmes for childhood cancer, this is significantly dire in emerging countries where 80 per cent of children diagnosed with cancer do not survive, while the reverse is the case in more developed societies. As a result, we are building ‘DUROTIMI’ a pre-diagnostic AI tool to aid early detection of cancer among ‘routine’ patients, thereby empowering physicians with the expertise of an oncologist with the aim of identifying and triaging high-risk patients early in the process.
Reason for Nomination:
Although there is an official staging sys tem and guidelines to classify early and advanced cancer stage, the terminology of ‘early’ and ‘advanced’ regarding cancer signs and symptoms is not precise and is used differently by health care providers and in different countries. There are numerous mechanisms in place in the medical domain that help in the early detection of cancer or other ailments. This majorly consists of 2 components i.e., early diagnosis (or downstaging) and screening. Early diagnosis focuses on detecting symptomatic patients as early as possible, while screening consists of testing healthy individuals to identify those having cancers before any symptoms appear. As per the most recent WHO report, 80% of countries do not have such early detection pro grammes/guidelines for detecting childhood cancer.
According to WHO, there were over 18 million cases of cancer in 2018 and over 9.5 million deaths from cancer. While cancer afflicts people of all ages, gender and demographic spectrum, in high-income countries, more than 80 per cent of children with cancers are cured, but this figure is only 20 per cent in many lo w and middle-income countries. The disparity is mostly the result of a late or inaccurate diagnosis, prohibitive treatment costs and insufficiently equipped facilities and trained medical personnel. Many cancer patients could be saved from premature death and suffering if they had timely access to
early detection programmes and adequate treatment.
After thorough research and consultation with physicians, radiologists, oncologists, and other specialists in this field, Radmol AI aims to deploy an AI-driven early detection platform for childhood cancer, DUROTIMI.
A significantly large amount of data is fed into Radmol AI’s Durotimi solution, which uses Natural Language Processing and Deep Learning to classify the symptoms in two triage levels, high or low. This solution combines a speech recognition module, a historical symptom-ailment repository, and a proprietary AI-enabled app.
Beyond the initial scope of addressing the risk of childhood cancer, this solution will enable physicians to identify other forms of cancer before they become obvious, thereby helping people live longer by increasing their odds of survival.