Hurone AI, Inc, Seattle, WA
Kingsley Ndoh , Christian Ruchaho Ntizimira , Irene Dankwa-Mullan , Theoneste Maniragaba
Background: The growing cancer burden in Africa is a major public health challenge. There were 8,944 new cases in Rwanda, with over 6,044 deaths reported in 2020. Unfortunately, there are only 13 oncologists in Rwanda for a population of 13 million people. The use of artificial intelligence (AI) tools and digital health in cancer treatment has the potential to mitigate the oncology workforce shortages, support clinical decision-making, and increase access to care. Hurone AI is a Seattle MedTech startup building culturally sensitive AI applications to address the oncologist-patient gaps and improve drug safety in underserved populations. Hurone’s premier application, Gukiza, launched at the Rwanda Cancer Center in 2022 to conduct beta testing. Gukiza is a remote patient monitoring system that ensures patients can report side effects or symptoms from their phones and oncology care teams can get real-time treatment insights and provide timely and effective care. Methods: Hurone received ethical approval from the Rwanda National Ethical Committee to launch a pilot in September 2022. An initial 45 breast cancer patients aged 20 – 65 who were either newly diagnosed or on active treatment were recruited. Each patient periodically received prompts that asked them about side effects and to score the degree of severity. The questions were adapted from the NCI’s adverse events repository. Gukiza analyzed each response and presented a visual analysis of each patient, enabling the cancer care team to send text-based interventions to the patient’s phone. Potential emergencies are flagged and sent emergency numbers to call. Through Amazon’s cloud analytic tools, Gukiza provides treatment insights to support oncologists' clinical decisions. Results: Results are summarized. Conclusions: Resource-appropriate digital technologies can be a useful tool in mitigating adverse events during cancer treatment and increasing access to timely care for patients. The data built in such systems can be a useful resource to individualize and improve cancer care and treatment outcomes.
User profile | Frequency | |
---|---|---|
Clinical oncologists | 3 | |
Nurses | 4 | |
Patients | 45 | |
Breast Cancer Types and Stages | ||
Diagnosis (breast cancer) | Frequency | Percentage |
Infiltrative ductal carcinoma | 30 | 75% |
Lobular carcinoma | 7 | 18% |
Mucinous | 3 | 8% |
Inflammatory breast cancer | 1 | 3% |
Others | 1 | 3% |
Cancer stage | Frequency | Percentage |
---|---|---|
Stage I | 1 | 2% |
Stage 2 | 24 | 53% |
Stage 3 | 19 | 42% |
Stage 4 | 1 | 2% |
User Engagement | Frequency/Percentages | |
Number of prompts | 2,825 | |
Number of patient responses | 2,690 | |
Number of interventions sent | 158 | |
Number of automatic emergency interventions | 9 | |
% of nurse-led interventions | 78% | |
% of physician-led interventions | 22% |
Disclaimer
This material on this page is ©2024 American Society of Clinical Oncology, all rights reserved. Licensing available upon request. For more information, please contact licensing@asco.org
Abstract Disclosures
2023 ASCO Annual Meeting
First Author: Natansh Modi
2022 ASCO Annual Meeting
First Author: Xiuwen Guan
2023 ASCO Annual Meeting
First Author: Kristen E. Champion
2022 ASCO Genitourinary Cancers Symposium
First Author: Jones T. Nauseef