Innovation: Patient-Reported Outcome/Side-Effects (PROSE) is a cloud-based personalized digital health tool for cancer patients and hospitals. It is an electronic Patient-Reported Outcome (ePRO) solution to monitor and follow-up patients experiencing treatment related side effects (TRSEs). It features an electronic health record and symptom tool that progressively captures real-world data of patient-reported outcomes (text, audio and image). PROSE is our proprietary health data science platform developed by Oncopadi researchers, patients and specialists with funding from the American Society of Clinical Oncology Innovation grant. PROSE addresses daily TRSE challenges: under-reporting, lack of clinical support/ advice and high morbidity. PROSE ML clinical algorithm will assess, track, predict and advice patients on treatment-related side effects While cancer specialists have unstructured and disorganized data that do not generate clinical or research insights, PROSE is a cost-effective technology increasing value in cancer care with these benefits; Real-time communication system and follow-up - Clinical advice and Patient education for self-care. Remote management of side effects (Acute, sub-acute and late). -Alert to physically report to clinic. Real-world data -Structured data collection and analysis for TRSE. -Utilize data for research and improve clinical processes. Support during non-clinic hours. - Stay connected with care team. - Make confident care decisions when experiencing TRSEs

Sector/Industry Application: Health: mHealth research, Digital/ Tele-Oncology.

Description: Head and neck cancers (HNC) are not uncommon in Nigeria with the prognosis largely poor due to late presentation to the hospital for prompt diagnosis and treatment. With the required aggressive multImodal treatments, patients experience debilitating symptoms and a significant burden of side effects. Close symptom monitoring throughout the care pathway can identify patients at risk of treatment dropout and poor quality of life and may assist in identifying patients who require medical assistance. The application of technologies for cancer care has the potential to guide big data approaches, such as those that seek to predict patients at risk of high treatment-related toxicity, poor treatment outcomes and side effect profiles, which is undoubtedly a step toward personalised cancer care. Machine learning (ML), an arm of Artificial Intelligence (AI) that uses big data to predict patterns and develop clinical algorithms, could potentially give radiation oncologists precise models to improve the care pathways for HNC patients. We, therefore, aim to develop a machine learning model that would profile, prognosticate and predict HNC patients to inform risk stratification and preventive strategies. An earlier PROSE pilot study, recruited 137 cancer patients receiving radiotherapy from March-October 2022 at NSIA-LUTH Cancer Centre to capture over 45,000 treatment-related side effects datasets. Using several supervised machine learning models, including Support Vector Machines, regression models, and Neural Networks, will be deployed to develop the predictive clinical algorithm. In conclusion, Patient-reported outcome data and research insights from the PROSE pilot study has led to exploring new research directions, one of which is to incorporate an Artificial intelligence (AI) clinical algorithm on the PROSE app. The AI-enabled web-based application will provide real-time self-care advice for patients experiencing moderate to severe side effects and predict patients at a high risk of dropping out of treatment due to side effects.

Problem: All cancer treatments cause side effects. Despite time spent in the hospital during active cancer care, many treatment-related side effects (TRSE) and symptom burdens occur outside the hospital. These are often under-reported and exacerbated by the shortage of clinical oncologists. Nigeria with a population of over 200 million people has only 82 clinical oncologists. The ratio of a clinical oncologist to new cancer patients is 1:1,500. It is even worse for 18 out of 36 states that do not have a clinical oncologist. This shortage has resulted in patients with TRSE being stranded with no clinical support, uninformed self-medication, and poor management of these TRSEs with a resultant reduction in quality of life. Traditionally, TRSEs are reported and managed at face-to-face consultations in the side effect clinic. Whilst this has its benefits, it poses several inconveniences to the already overwhelmed cancer patient who has less than an hour out of 720 hours a month to access the oncologist. It has been projected that capturing patient-reported health data, including reporting TRSEs in real-time, is of utmost importance in allowing rapid clinical decision-making and intervention. The limited options for real-time reporting of TRSEs and prompt intervention give rise to approaches that can facilitate and integrate reports into routine clinical practice. By delivering the most important instructions at the right time, PROSE improves treatment outcomes, keeps patients safe, and improves patient compliance - therefore optimizing specialists' time spent and avoiding unnecessary delays.