Evaluation of techniques to improve a deep learning algorithm for the automatic detection of intracranial haemorrhage on CT head imaging

This study evaluates deep learning (DL) algorithms that are playing an increasingly important role in automatic medical image analysis. The DL algorithm used was trained and externally evaluated on open-source, multi-centre retrospective data that contained radiologist-annotated non-contrast CT head studies. The authors concluded that the DL model has applications in a triage role with the potential to improve diagnostic yield and efficiency of CT reporting, resulting in expediting treatment and improving outcomes for intracranial haemorrhage.

Key points

• The deep learning model detected intracranial haemorrhages on computed tomography with high accuracy.
• Image preprocessing, such as windowing, plays a large role in improving deep learning model performance.
• Implementations which enable an analysis of interslice dependencies can improve deep learning model performance.
• Visual saliency maps can facilitate explainable artificial intelligence systems.
• Deep learning within a triage system may expedite earlier intracranial haemorrhage detection.

Article: Evaluation of techniques to improve a deep learning algorithm for the automatic detection of intracranial haemorrhage on CT head imaging

Authors: Melissa Yeo, Bahman Tahayori, Hong Kuan Kok, Julian Maingard, Numan Kutaiba, Jeremy Russell, Vincent Thijs, Ashu Jhamb, Ronil V. Chandra, Mark Brooks, Christen D. Barras & Hamed Asadi