Based on First Nations health statistics in remote communities, Northern Territory reveals its third-world country reality in a first-world nation. We choose the insufficient antenatal care in remote communities as an entry point to the haunting structural problem, where over 30% of First Nations pregnant women in communities are under 20 years old, with over 50% are smokers and most of them are inaccessible to digital devices. They have 5 times less medical support than pregnant women in urban areas and less chance to access medicine, needless to imagine early child education and preparation.
Due to the diverse conditions in different remote communities, we are proposing a trial project to a community where we can work with elders. We are honored to have Auntie Crystal Love joining our GovHack team to re-imagine antenatal care on Tiwi Islands. Auntie Crystal verifies the open data and listed out more problems hidden away from statistical stories, including youth drug abuse, alchoholicism, sexual transmission disease, dialysis due to mutation, distrust for medical authorities and high suicide rate. Auntie Crystal also indicates clinical environments on Tiwi are lack of traditional protocol of gender and skin kinship (Wantarringuwi, Miyartiwi, Marntimapila & Takaringuwi), which is another critical factor why young Tiwi pregnant women prefer care from elders than clinical support.
Our team proposes a medical assistant software that has two components. The first component is powered by Azure AI to translate medical diagnosis advice into Aboriginal language (Tiwi in our trial case) with a visual interface of autonomy to pinpoint discomfort. This component is designed with traditional protocol of specific communities. It is aimed to upskill community elders for antenatal care among their pregnant kins and improve technical literacy among female Aboriginal groups. It is compatible with multiple platforms and can be installed in different types of smart digital devices.
The second component is data collection for medical interoperability. Collected data from the device will improve AI models for natural language processing between Aboriginal and medical lexicon. It will also be processed by machine learning models by InterSystems, the same system used by NT Acacia Digital Health System, to provide advance prediction such as re-admission of smoker patients (see below demo). These predictions can ease medical burden and reduce traveling to hospital from remote communities. This will also reduce cost for Aboriginal interpreters and can train community elders to align with medical protocol to assist the health profession to deliver comprehensive services.
NT Health Acacia AI integration to predict smoker re-admission demo: https://8080-0-8bbd7b0b.labs.learning.intersystems.com/?fbclid=IwAR2uhCGNISFF6lWyAlyf8QtF_RAie1oa0XllSueltt5aucOtc0f15KkkXcQ#/notebook/LESSON1