An Android-Based Mobile App (ARVPredictor) for the Detection of HIV Drug-Resistance Mutations and Treatment at the Point of Care: Development Study
Bibliographic record
Abstract
BACKGROUND: HIV/AIDS remains one of the major global human health challenges, especially in resource-limited environments. By 2017, over 77.3 million people were infected with the disease, and approximately 35.4 million individuals had already died from AIDS-related illnesses. Approximately 21.7 million people were accessing ART with significant clinical outcomes. However, numerous challenges are experienced in the delivery and accurate interpretation of data on patients with HIV data by various health care providers at different care levels. Mobile health (mHealth) technology is progressively making inroads into the health sector as well as medical research. Different mobile devices have become common in health care settings, leading to rapid growth in the development of downloadable software specifically designed to fulfill particular health-related purposes. OBJECTIVE: We developed a mobile-based app called ARVPredictor and demonstrated that it can accurately define HIV-1 drug-resistance mutations in the HIV pol gene for use at the point of care. METHODS: ARVPredictor was designed using Android Studio with Java as the programming language and is compatible with both Android and iOS. The app system is hosted on Nginx Server, and network calls are built on PHP's Laravel framework handled by the Retrofit Library. The DigitalOcean offers a high-performance and stable cloud computing platform for ARVPredictor. This mobile app is enlisted in the Google Play Store as an "ARVPredictor" and the source code is available under MIT permissive license at a GitHub repository. To test for agreement between the ARVPredictor and Stanford HIV Database in detecting HIV subtype and NNRT and NRTI mutations, a total of 100 known HIV sequences were evaluated. RESULTS: The mobile-based app (ARVPredictor) takes in a set of sequences or known mutations (protease, reverse transcriptase and integrase). It then returns inferred levels of resistance to selected nucleoside, nonnucleoside protease, and integrase inhibitors for accurate HIV/AIDS management at the point of care. The ARVPredictor identified similar HIV subtypes in 98/100 sequences compared with the Stanford HIV Database (κ=0.98, indicating near perfect agreement). There were 89/100 major NNRTI and NRTI mutations identified by ARVPredictor, similar to the Stanford HIV Database (κ=0.89, indicating near perfect agreement). Eight mutations classified as major by the Stanford HIV Database were classified as others by ARVPredictor. CONCLUSIONS: The ARVPredictor largely agrees with the Stanford HIV Database in identifying both major and minor proteases, reverse transcriptase, and integrase mutations. The app can be conveniently used robustly at the point of care by HIV/AIDS care providers to improve the management of HIV infection.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".