Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

Automated detection of underdiagnosed medical conditions via opportunistic imaging

Created by
  • Haebom

Author

Asad Aali, Andrew Johnston, Louis Blankemeier, Dave Van Veen, Laura T Derry, David Svec, Jason Hom, Robert D. Boutin, Akshay S. Chaudhari

Outline

This study evaluated the potential of opportunistic computed tomography (CT) for diagnosing underdiagnosed conditions such as sarcopenia, fatty liver, and ascites using a deep learning method. We analyzed 2,674 inpatient CT scans to identify discrepancies between imaging phenotypes derived from opportunistic CT scans and radiology reports and International Classification of Diseases (ICD) coding. We found that only 0.5%, 3.2%, and 30.7% of sarcopenia, fatty liver, and ascites diagnosed through opportunistic imaging or radiology reports, respectively, were recorded with ICD codes. This suggests that opportunistic CT can contribute to the advancement of precision medicine by improving diagnostic accuracy and the accuracy of risk-adjustment models.

Takeaways, Limitations

•
Takeaways:
◦
We demonstrate that opportunistic CT can improve the diagnostic accuracy of underdiagnosed conditions such as sarcopenia, fatty liver, and ascites.
◦
These findings suggest that opportunistic CT may contribute to the advancement of precision medicine by improving diagnostic accuracy and enhancing the accuracy of risk-adjustment models.
◦
It suggests the need to develop measures to improve the coding accuracy of medical information systems.
•
Limitations:
◦
Because this study was conducted based on inpatient data from a specific hospital, generalizability to other settings may be limited.
◦
There is a lack of information about the specifics of the deep learning model used in the analysis and the performance evaluation metrics.
◦
There is a lack of in-depth analysis of the causes of ICD coding omissions. Only the omission rate is presented, without any analysis of the underlying causes.
👍
Made with Slashpage