AI platform using small data sets improves treatment response

While artificial intelligence (AI) traditionally taps on big data from a large patient population, the promise of AI for healthcare becomes severely limited when it comes to patients suffering from rare diseases – classified as fewer than 1 in 2,000 people being afflicted.

To overcome this handicap, the National University of Singapore’s School of Medicine (NUS Medicine) has demonstrated in one trial that using AI on the individual patient’s clinical data to personalise treatments has shown significant treatment outcomes.

NUS Medicine’s CURATE.AI platform dynamically adjusts treatment doses based on the patient’s responses.

To subscribe to the GovInsider bulletin click here. 

Reduced dose and improved patient outcomes

Over a two-year period, a patient suffering from a rare blood disorder showed treatment response improvement, fewer hospital admissions, and healthcare cost savings.

These included substantial improvement in red blood cell levels, thereby being able to avoid blood transfusions, and no serious side effects from treatment. Notably, the patient saved approximately S$10,500 on drug costs.

“No two patients are alike, and even the same patient can change over time as well. It is essential for treatment to evolve alongside the patient.

“Our study highlights the effectiveness of using small data to treat extremely rare diseases— addressing the gaps where traditional big data methods fall short, and where largescale trials are not feasible due to the limited patient population,” said the research co-lead, NUS Medicine’s Director of the Institute for Digital Medicine (WisDM), Professor Dean Ho.

Previous studies involving different patient groups also saw improved treatment responses. Interestingly, the platform also recommended lower doses compared to those prescribed by the doctors.

One was in NUS Medicine’s collaboration with a US-based hospital where a patient with advanced prostate cancer resumed an active lifestyle after the drug dose was reduced by 50 per cent and deemed more tolerable.

Another patient in Singapore suffering from advanced cancer saw his lung tumour shrunk and cancer progression stopped after he was recommended a reduced dose.

To subscribe to the GovInsider bulletin click here.  

Human in the loop

For this trial, NUS Medicine worked with clinicians from the National University Cancer Institute, Singapore (NCIS) to accept or reject the doses recommended by the platform.

The very first clinical studies conducted around the platform’s use were reported by The Straits Times two years ago, where NUS Medicine and NCIS partnered on a trial for patients with advanced solid tumours.

The study back then found that clinicians accepted close to 97 per cent of the recommendations, and most of the doses were around 20 per cent lower than that prescribed by clinicians.

NCIS’ Department of Hematology-Oncology, Dr Raghav Sundar, previously in The Straits Times article also underlined the key benefit of this platform to enable clinicians to focus more on the patient and caregiver, than just focusing on the patient's treatment.

The trial is currently still ongoing and is open to recruitment of new, suitable patients. The results collected from the first two years have been published in the NPJ Digital Medicine journal.