Publication on deep learning-based intracranial hemorrhage assessment in non-contrast CT scans

DEI is happy to share our newly published study on deep learning-based intracranial hemorrhage (ICH) assessment in non-contrast CT scans published in La Radiologia Medica.

Study Highlights:
Diagnostic Accuracy: The deep-learning algorithm achieved an accuracy of 91.24% in identifying ICH, with a sensitivity of 90.37% and specificity of 94.85%. For distinguishing between different types of hemorrhage, it reached an accuracy of 98.54%.
Rapid Response: The software delivered diagnoses in just 15 seconds on average
Comprehensive Detection: It classified various types of ICH, including intraparenchymal, intraventricular, subarachnoid, subdural, and epidural hemorrhages, and detected midline shifts with 100% sensitivity.

What This Means for Clinical Routine:
Improved Patient Care: By delivering rapid, accurate results, the algorithm supports better patient care, minimizing the risks associated with delayed or incorrect diagnoses.
Workload: Radiologists working overnight often face high volumes and fatigue. This algorithm provides reliable initial assessments, potentially allowing them to focus on critical decision-making.

For more information please click on the link below

Accuracy and time efficiency of a novel deep learning algorithm for Intracranial Hemorrhage detection in CT Scans | La radiologia medica (springer.com)