1
|
Yeo M, Tahayori B, Kok HK, Maingard J, Kutaiba N, Russell J, Thijs V, Jhamb A, Chandra RV, Brooks M, Barras CD, Asadi H. Evaluation of techniques to improve a deep learning algorithm for the automatic detection of intracranial haemorrhage on CT head imaging. Eur Radiol Exp 2023; 7:17. [PMID: 37032417 PMCID: PMC10083149 DOI: 10.1186/s41747-023-00330-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/09/2022] [Accepted: 02/07/2023] [Indexed: 04/11/2023] Open
Abstract
BACKGROUND Deep learning (DL) algorithms are playing an increasing role in automatic medical image analysis. PURPOSE To evaluate the performance of a DL model for the automatic detection of intracranial haemorrhage and its subtypes on non-contrast CT (NCCT) head studies and to compare the effects of various preprocessing and model design implementations. METHODS The DL algorithm was trained and externally validated on open-source, multi-centre retrospective data containing radiologist-annotated NCCT head studies. The training dataset was sourced from four research institutions across Canada, the USA and Brazil. The test dataset was sourced from a research centre in India. A convolutional neural network (CNN) was used, with its performance compared against similar models with additional implementations: (1) a recurrent neural network (RNN) attached to the CNN, (2) preprocessed CT image-windowed inputs and (3) preprocessed CT image-concatenated inputs. The area under the receiver operating characteristic curve (AUC-ROC) and microaveraged precision (mAP) score were used to evaluate and compare model performances. RESULTS The training and test datasets contained 21,744 and 491 NCCT head studies, respectively, with 8,882 (40.8%) and 205 (41.8%) positive for intracranial haemorrhage. Implementation of preprocessing techniques and the CNN-RNN framework increased mAP from 0.77 to 0.93 and increased AUC-ROC [95% confidence intervals] from 0.854 [0.816-0.889] to 0.966 [0.951-0.980] (p-value = 3.91 × 10-12). CONCLUSIONS The deep learning model accurately detected intracranial haemorrhage and improved in performance following specific implementation techniques, demonstrating clinical potential as a decision support tool and an automated system to improve radiologist workflow efficiency. 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.
Collapse
Affiliation(s)
- Melissa Yeo
- Melbourne Medical School, The University of Melbourne, Melbourne, VIC, Australia.
| | - Bahman Tahayori
- Department of Biomedical Engineering, The University of Melbourne, Melbourne, VIC, Australia
- IBM Research Australia, Melbourne, VIC, Australia
| | - Hong Kuan Kok
- Interventional Radiology Service, Department of Radiology, Northern Health, Epping, VIC, Australia
- School of Medicine, Faculty of Health, Deakin University, Burwood, VIC, Australia
| | - Julian Maingard
- School of Medicine, Faculty of Health, Deakin University, Burwood, VIC, Australia
- Interventional Neuroradiology Unit, Monash Health, Clayton, VIC, Australia
- Faculty of Medicine Nursing and Health Sciences, Monash University, Clayton, VIC, Australia
- Department of Radiology, St Vincent's Hospital, Melbourne, VIC, Australia
| | - Numan Kutaiba
- Department of Radiology, Austin Hospital, Melbourne, VIC, Australia
| | - Jeremy Russell
- Department of Neurosurgery, Austin Hospital, Melbourne, VIC, Australia
| | - Vincent Thijs
- Stroke Theme, Florey Institute of Neuroscience and Mental Health, Melbourne, VIC, Australia
- Department of Neurology, Austin Health, Melbourne, VIC, Australia
| | - Ashu Jhamb
- Department of Radiology, St Vincent's Hospital, Melbourne, VIC, Australia
| | - Ronil V Chandra
- Interventional Neuroradiology Unit, Monash Health, Clayton, VIC, Australia
- Faculty of Medicine Nursing and Health Sciences, Monash University, Clayton, VIC, Australia
| | - Mark Brooks
- Melbourne Medical School, The University of Melbourne, Melbourne, VIC, Australia
- School of Medicine, Faculty of Health, Deakin University, Burwood, VIC, Australia
- Stroke Theme, Florey Institute of Neuroscience and Mental Health, Melbourne, VIC, Australia
- Interventional Neuroradiology Service, Department of Radiology, Austin Hospital, Melbourne, VIC, Australia
| | - Christen D Barras
- South Australian Institute of Health and Medical Research, Adelaide, South Australia, Australia
- School of Medicine, The University of Adelaide, Adelaide, South Australia, Australia
| | - Hamed Asadi
- Melbourne Medical School, The University of Melbourne, Melbourne, VIC, Australia
- School of Medicine, Faculty of Health, Deakin University, Burwood, VIC, Australia
- Interventional Neuroradiology Unit, Monash Health, Clayton, VIC, Australia
- Department of Radiology, St Vincent's Hospital, Melbourne, VIC, Australia
- Stroke Theme, Florey Institute of Neuroscience and Mental Health, Melbourne, VIC, Australia
- Interventional Neuroradiology Service, Department of Radiology, Austin Hospital, Melbourne, VIC, Australia
| |
Collapse
|
2
|
Loizidou K, Skouroumouni G, Pitris C, Nikolaou C. Digital subtraction of temporally sequential mammograms for improved detection and classification of microcalcifications. Eur Radiol Exp 2021; 5:40. [PMID: 34519867 PMCID: PMC8440760 DOI: 10.1186/s41747-021-00238-w] [Citation(s) in RCA: 7] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/30/2021] [Accepted: 08/04/2021] [Indexed: 11/24/2022] Open
Abstract
Background Our aim was to demonstrate that automated detection and classification of breast microcalcifications, according to Breast Imaging Reporting and Data System (BI-RADS) categorisation, can be improved with the subtraction of sequential mammograms as opposed to using the most recent image only. Methods One hundred pairs of mammograms were retrospectively collected from two temporally sequential rounds. Fifty percent of the images included no (BI-RADS 1) or benign (BI-RADS 2) microcalcifications. The remaining exhibited suspicious findings (BI-RADS 4-5) in the recent image. Mammograms cannot be directly subtracted, due to tissue changes over time and breast deformation during mammography. To overcome this challenge, optimised preprocessing, image registration, and postprocessing procedures were developed. Machine learning techniques were employed to eliminate false positives (normal tissue misclassified as microcalcifications) and to classify the true microcalcifications as BI-RADS benign or suspicious. Ninety-six features were extracted and nine classifiers were evaluated with and without temporal subtraction. The performance was assessed by measuring sensitivity, specificity, accuracy, and area under the curve (AUC) at receiver operator characteristics analysis. Results Using temporal subtraction, the contrast ratio improved ~ 57 times compared to the most recent mammograms, enhancing the detection of the radiologic changes. Classifying as BI-RADS benign versus suspicious microcalcifications, resulted in 90.3% accuracy and 0.87 AUC, compared to 82.7% and 0.81 using just the most recent mammogram (p = 0.003). Conclusion Compared to using the most recent mammogram alone, temporal subtraction is more effective in the microcalcifications detection and classification and may play a role in automated diagnosis systems. Supplementary Information The online version contains supplementary material available at 10.1186/s41747-021-00238-w.
Collapse
Affiliation(s)
- Kosmia Loizidou
- KIOS Research and Innovation Center of Excellence, Department of Electrical and Computer Engineering, University of Cyprus, 1 Panepistimiou Avenue, Aglantzia, 2109, Nicosia, Cyprus.
| | - Galateia Skouroumouni
- Nicosia General Hospital, 215 Nicosia-Limassol Old Road, Strovolos, 2029, Nicosia, Cyprus
| | - Costas Pitris
- KIOS Research and Innovation Center of Excellence, Department of Electrical and Computer Engineering, University of Cyprus, 1 Panepistimiou Avenue, Aglantzia, 2109, Nicosia, Cyprus
| | | |
Collapse
|