"Informed AI News" is an publications aggregation platform, ensuring you only gain the most valuable information, to eliminate information asymmetry and break through the limits of information cocoons. Find out more >>
Google's AI Advances in Early Disease Detection Through Bioacoustics
- summary
- score
Google's AI, HeAR, listens for sickness. It utilizes bioacoustics, merging biology with sound. This technology predicts diseases early, which is particularly useful in regions with inadequate healthcare.
HeAR is trained on 300 million audio clips—ranging from coughs, sniffles, sneezes, to breaths. The data is sourced from YouTube, using non-copyrighted material. One notable dataset comes from Zambian hospital patients screened for TB. HeAR excels in identifying coughs, potentially spotting TB early.
In collaboration with Salcit Technologies, an Indian AI healthcare startup, HeAR's capabilities are enhanced. Salcit's Swaasa app collects cough samples and diagnoses diseases with 94% accuracy, costing just $2.40—a cheaper alternative to traditional tests.
Challenges include background noise in samples, and HeAR is not yet ready for the market. Nonetheless, the concept of using AI for sound-based diagnosis holds promise.
| Scores | Value | Explanation |
|---|---|---|
| Objectivity | 6 | Balanced reporting with comprehensive analysis. |
| Social Impact | 5 | Significantly influences public opinion on healthcare. |
| Credibility | 5 | Solid evidence from authoritative sources. |
| Potential | 6 | High potential to lead to significant healthcare changes. |
| Practicality | 5 | Extremely practical, widely applied in practice. |
| Entertainment Value | 2 | Slightly monotonous with a few entertaining elements. |