External validation study on the value of deep learning algorithm for the prediction of hematoma expansion from noncontrast CT scans

Abstract

Background Hematoma expansion is an independent predictor of patient outcome and mortality. The early diagnosis of hematoma expansion is crucial for selecting clinical treatment options. This study aims to explore the value of a deep learning algorithm for the prediction of hematoma expansion from non-contrast computed tomography (NCCT) scan through external validation. Methods 102 NCCT images of hypertensive intracerebral hemorrhage (HICH) patients diagnosed in our hospital were retrospectively reviewed. The initial computed tomography (CT) scan images were evaluated by a commercial Artificial Intelligence (AI) software using deep learning algorithm and radiologists respectively to predict hematoma expansion and the corresponding sensitivity, specificity and accuracy of the two groups were calculated and compared. Comparisons were also conducted among gold standard hematoma expansion diagnosis time, AI software diagnosis time and doctors’ reading time. Results Among 102 HICH patients, the sensitivity, specificity, and accuracy of hematoma expansion prediction in the AI group were higher than those in the doctor group(80.0% vs 66.7%, 73.6% vs 58.3%, 75.5% vs 60.8%), with statistically significant difference (p < 0.05). The AI diagnosis time (2.8 ± 0.3 s) and the doctors’ diagnosis time (11.7 ± 0.3 s) were both significantly shorter than the gold standard diagnosis time (14.5 ± 8.8 h) (p < 0.05), AI diagnosis time was significantly shorter than that of doctors (p < 0.05). Conclusions Deep learning algorithm could effectively predict hematoma expansion at an early stage from the initial CT scan images of HICH patients after onset with high sensitivity and specificity and greatly shortened diagnosis time, which provides a new, accurate, easy-to-use and fast method for the early prediction of hematoma expansion.

Prediction of hematoma expansion in spontaneous intracerebral hemorrhage using a multimodal neural network

“…The voxel size of the CT images was 0.5 × 0.5 × 2.0 mm, where the kernel size of 3 or 5 may have been too small to extract features from the hematoma. The larger kernel sizes up to 19 × 19 × 7 worked effectively in this study; we could not confirm the kernel size in other studies that used CNN to predict hematoma expansion because the programming codes were not disclosed 18 – 21 , 29 .…”

Research advances in predicting the expansion of hypertensive intracerebral hemorrhage based on CT images: an overview

“… Note: Lee et al (2024) , Wu et al (2024) , Tran et al (2024) , Bo et al (2023) , Feng et al (2023) , Tang et al (2022) , Guo et al (2022) , Ma & Zhou (2022) , Zhong et al (2021) , Teng et al (2021) . …”

“…Evaluating its safety and accuracy is currently a hot research topic. Guo et al (2022) utilized commercial artificial intelligence (AI) software to assess initial CT scan images, predict hematoma expansion using both a deep learning algorithm and a radiologist, and calculate/compare the corresponding sensitivity, specificity, and accuracy of the two groups. They compared the gold standard diagnostic time for hematoma dilation with the diagnostic time of artificial intelligence software and doctors’ reading time.…”