Abstract 408P
Background
Glioblastoma (GBM) is one of the most common malignant primary brain tumours, which accounts for 60-70% of all glioma. Conventional diagnosis and post-operation treatment plan for glioblastoma is mainly based on the identification of tumor cells on haematoxylin and eosin stained (H&E) histopathological slides by both experienced medical technologist and pathologist. The recent development of digital whole slide scanners allows the histopathological image analysis feasible. This study aimed to build an image feature based model using histopathology whole slide images to differentiate patients with GBM from healthy control (HC).
Methods
628 H&E images were collected from 30 GBM patients, obtained from public database the Cancer Image Archive. Another 554 images were collected from 30 HC. 22 Gray level Co-occurrence Matrix (GLCM) image features were retrieved from the collected images by an in-house developed software. Both the GBM and HC groups were divided into a training and a validation group on 70:30 basis. A random forest algorithm was used to build the model using training set, and validated by the validation set by R software.
Results
The receiver operating characteristics (ROC) curve showed that the average class and overall accuracy achieved 94.6%. The area under the curve (AUC) was 0.98, with sensitivity of 94.1% and sensitivity of 95.1%.
Conclusions
The machine learning model can accurately differentiate GBM from HC in histopathology whole slide imaging. The application can be fully automatic and it is practical in applying for GBM screening.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
College Research Grant (CRG) - Tung Wah College.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
399P - Role of denosumab as targeted therapy in treatment of unresectable or recurrent cases of giant cell tumor of the bone
Presenter: Anchal Mishra
Session: Poster viewing 06
400P - Phase I study of fianlimab, a human lymphocyte activation gene-3 (LAG-3) monoclonal antibody, in combination with cemiplimab in advanced melanoma (mel)
Presenter: Omid Hamid
Session: Poster viewing 06
401P - A phase Ib study of E7386, a CREB-binding protein (CBP)/β-catenin interaction inhibitor, in combination with lenvatinib in patients (pts) with advanced solid tumors
Presenter: Shunsuke Kondo
Session: Poster viewing 06
402P - Phase I study of selumetinib in Chinese pediatric and adult patients (pts) with neurofibromatosis type 1 (NF1) and inoperable plexiform neurofibromas (PN): Interim results
Presenter: Xiao-Jun Yuan
Session: Poster viewing 06
403P - Gender trend over time of principal investigators within large Australian cancer clinical trials groups
Presenter: Thi Thao Vi Luong
Session: Poster viewing 06
404P - Adolescent & young adult cancers (AYA): Experience from a large rural teaching hospital in South India
Presenter: Vishwapriya Godkhindi
Session: Poster viewing 06
405P - Evaluating medical oncology outcomes (EMOO) in Asia study: Diagnosis, staging, treatment patterns and outcome of patients with lung cancer registered 2017-2019
Presenter: Ross Soo
Session: Poster viewing 06
406P - Adrenal insufficiency induced by immune checkpoint inhibitors: Clinical characteristics of 145 cases
Presenter: kaoru koshiba
Session: Poster viewing 06
407P - A post-marketing surveillance of the real-world safety and effectiveness of avelumab in patients with curatively unresectable Merkel cell carcinoma in Japan
Presenter: Naoya Yamazaki
Session: Poster viewing 06
409P - Tertiary lymphoid structures are scarce but associated with BCR clonal expansion, B cell activity and checkpoint inhibitor response in advanced osteosarcoma
Presenter: Qiyuan Bao
Session: Poster viewing 06