Summary of Discriminant Audio Properties in Deep Learning Based Respiratory Insufficiency Detection in Brazilian Portuguese, by Marcelo Matheus Gauy et al.
Discriminant audio properties in deep learning based respiratory insufficiency detection in Brazilian Portuguese
by Marcelo Matheus Gauy, Larissa Cristina Berti, Arnaldo Cândido Jr, Augusto Camargo Neto, Alfredo Goldman, Anna Sara Shafferman Levin, Marcus Martins, Beatriz Raposo de Medeiros, Marcelo Queiroz, Ester Cerdeira Sabino, Flaviane Romani Fernandes Svartman, Marcelo Finger
First submitted to arxiv on: 27 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Sound (cs.SD); Audio and Speech Processing (eess.AS)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Medium Difficulty summary: This paper explores AI systems that detect respiratory insufficiency (RI) by analyzing speech audios. Previous studies have demonstrated the feasibility of AI-based RI detection, achieving 96.5% accuracy using convolutional neural networks (CNNs) and Transformers. However, these models were trained on data from COVID-19 patients during the pandemic’s early phase. This work collects new RI patient data with various causes, including non-COVID-19 related cases, to extend AI-based RI detection. The study also includes control data from hospital patients without RI. Notably, the considered AI models, when trained on the initial COVID-19 data, fail to generalize to the new RI patient dataset, suggesting that COVID-19 RI has features not present in all RI types. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Low Difficulty summary: This research investigates how artificial intelligence (AI) can help diagnose respiratory problems by analyzing people’s voices. Previous studies have shown that AI can be very accurate in detecting these issues, but they only used data from patients with COVID-19. In this study, the researchers collect new data from patients with different types of respiratory problems to see if AI can still detect them accurately. They also compare the results to those from healthy people without respiratory problems. The surprising finding is that even though AI was very good at detecting COVID-19-related respiratory issues, it struggles when applied to other types of respiratory problems. |
