Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features
Background: Early differentiation of neoplastic and non-neoplastic intracerebral hemorrhage (ICH) can be difficult in initial radiological evaluation, especially for extensive ICHs. The aim of this study was to evaluate the potential of a machine learning-based prediction of etiology for acute ICHs...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2020-05-01
|
| Series: | Frontiers in Neurology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fneur.2020.00285/full |
| _version_ | 1828842612992245760 |
|---|---|
| author | Jawed Nawabi Helge Kniep Reza Kabiri Gabriel Broocks Tobias D. Faizy Christian Thaler Gerhard Schön Jens Fiehler Uta Hanning |
| author_facet | Jawed Nawabi Helge Kniep Reza Kabiri Gabriel Broocks Tobias D. Faizy Christian Thaler Gerhard Schön Jens Fiehler Uta Hanning |
| author_sort | Jawed Nawabi |
| collection | DOAJ |
| description | Background: Early differentiation of neoplastic and non-neoplastic intracerebral hemorrhage (ICH) can be difficult in initial radiological evaluation, especially for extensive ICHs. The aim of this study was to evaluate the potential of a machine learning-based prediction of etiology for acute ICHs based on quantitative radiomic image features extracted from initial non-contrast-enhanced computed tomography (NECT) brain scans.Methods: The analysis included NECT brain scans from 77 patients with acute ICH (n = 50 non-neoplastic, n = 27 neoplastic). Radiomic features including shape, histogram, and texture markers were extracted from non-, wavelet-, and log-sigma-filtered images using regions of interest of ICH and perihematomal edema (PHE). Six thousand and ninety quantitative predictors were evaluated utilizing random forest algorithms with five-fold model-external cross-validation. Model stability was assessed through comparative analysis of 10 randomly drawn cross-validation sets. Classifier performance was compared with predictions of two radiologists employing the Matthews correlation coefficient (MCC).Results: The receiver operating characteristic (ROC) area under the curve (AUC) of the test sets for predicting neoplastic vs. non-neoplastic ICHs was 0.89 [95% CI (0.70; 0.99); P < 0.001], and specificities and sensitivities reached >80%. Compared to the radiologists' predictions, the machine learning algorithm yielded equal or superior results for all evaluated metrics. The MCC of the proposed algorithm at its optimal operating point (0.69) was significantly higher than the MCC of the radiologist readers (0.54); P = 0.01.Conclusion: Evaluating quantitative features of acute NECT images in a machine learning algorithm provided high discriminatory power in predicting non-neoplastic vs. neoplastic ICHs. Utilized in the clinical routine, the proposed approach could improve patient care at low risk and costs. |
| first_indexed | 2024-12-12T20:28:05Z |
| format | Article |
| id | doaj.art-d68c82d9f227498abc09433aab9d9e58 |
| institution | Directory Open Access Journal |
| issn | 1664-2295 |
| language | English |
| last_indexed | 2024-12-12T20:28:05Z |
| publishDate | 2020-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurology |
| spelling | doaj.art-d68c82d9f227498abc09433aab9d9e582022-12-22T00:13:06ZengFrontiers Media S.A.Frontiers in Neurology1664-22952020-05-011110.3389/fneur.2020.00285518763Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image FeaturesJawed Nawabi0Helge Kniep1Reza Kabiri2Gabriel Broocks3Tobias D. Faizy4Christian Thaler5Gerhard Schön6Jens Fiehler7Uta Hanning8Department of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, GermanyDepartment of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, GermanyDepartment of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, GermanyDepartment of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, GermanyDepartment of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, GermanyDepartment of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, GermanyInstitute of Medical Biometry and Epidemiology, University Medical Center Hamburg-Eppendorf, Hamburg, GermanyDepartment of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, GermanyDepartment of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, GermanyBackground: Early differentiation of neoplastic and non-neoplastic intracerebral hemorrhage (ICH) can be difficult in initial radiological evaluation, especially for extensive ICHs. The aim of this study was to evaluate the potential of a machine learning-based prediction of etiology for acute ICHs based on quantitative radiomic image features extracted from initial non-contrast-enhanced computed tomography (NECT) brain scans.Methods: The analysis included NECT brain scans from 77 patients with acute ICH (n = 50 non-neoplastic, n = 27 neoplastic). Radiomic features including shape, histogram, and texture markers were extracted from non-, wavelet-, and log-sigma-filtered images using regions of interest of ICH and perihematomal edema (PHE). Six thousand and ninety quantitative predictors were evaluated utilizing random forest algorithms with five-fold model-external cross-validation. Model stability was assessed through comparative analysis of 10 randomly drawn cross-validation sets. Classifier performance was compared with predictions of two radiologists employing the Matthews correlation coefficient (MCC).Results: The receiver operating characteristic (ROC) area under the curve (AUC) of the test sets for predicting neoplastic vs. non-neoplastic ICHs was 0.89 [95% CI (0.70; 0.99); P < 0.001], and specificities and sensitivities reached >80%. Compared to the radiologists' predictions, the machine learning algorithm yielded equal or superior results for all evaluated metrics. The MCC of the proposed algorithm at its optimal operating point (0.69) was significantly higher than the MCC of the radiologist readers (0.54); P = 0.01.Conclusion: Evaluating quantitative features of acute NECT images in a machine learning algorithm provided high discriminatory power in predicting non-neoplastic vs. neoplastic ICHs. Utilized in the clinical routine, the proposed approach could improve patient care at low risk and costs.https://www.frontiersin.org/article/10.3389/fneur.2020.00285/fullintracerebral hemorrhageneoplastic hemorrhageradiomicsmachine learningartificial intelligence |
| spellingShingle | Jawed Nawabi Helge Kniep Reza Kabiri Gabriel Broocks Tobias D. Faizy Christian Thaler Gerhard Schön Jens Fiehler Uta Hanning Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features Frontiers in Neurology intracerebral hemorrhage neoplastic hemorrhage radiomics machine learning artificial intelligence |
| title | Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features |
| title_full | Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features |
| title_fullStr | Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features |
| title_full_unstemmed | Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features |
| title_short | Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features |
| title_sort | neoplastic and non neoplastic acute intracerebral hemorrhage in ct brain scans machine learning based prediction using radiomic image features |
| topic | intracerebral hemorrhage neoplastic hemorrhage radiomics machine learning artificial intelligence |
| url | https://www.frontiersin.org/article/10.3389/fneur.2020.00285/full |
| work_keys_str_mv | AT jawednawabi neoplasticandnonneoplasticacuteintracerebralhemorrhageinctbrainscansmachinelearningbasedpredictionusingradiomicimagefeatures AT helgekniep neoplasticandnonneoplasticacuteintracerebralhemorrhageinctbrainscansmachinelearningbasedpredictionusingradiomicimagefeatures AT rezakabiri neoplasticandnonneoplasticacuteintracerebralhemorrhageinctbrainscansmachinelearningbasedpredictionusingradiomicimagefeatures AT gabrielbroocks neoplasticandnonneoplasticacuteintracerebralhemorrhageinctbrainscansmachinelearningbasedpredictionusingradiomicimagefeatures AT tobiasdfaizy neoplasticandnonneoplasticacuteintracerebralhemorrhageinctbrainscansmachinelearningbasedpredictionusingradiomicimagefeatures AT christianthaler neoplasticandnonneoplasticacuteintracerebralhemorrhageinctbrainscansmachinelearningbasedpredictionusingradiomicimagefeatures AT gerhardschon neoplasticandnonneoplasticacuteintracerebralhemorrhageinctbrainscansmachinelearningbasedpredictionusingradiomicimagefeatures AT jensfiehler neoplasticandnonneoplasticacuteintracerebralhemorrhageinctbrainscansmachinelearningbasedpredictionusingradiomicimagefeatures AT utahanning neoplasticandnonneoplasticacuteintracerebralhemorrhageinctbrainscansmachinelearningbasedpredictionusingradiomicimagefeatures |