Machine Learning of Multi-Modal Tumor Imaging Reveals Trajectories of Response to Precision Treatment
The standard assessment of response to cancer treatments is based on gross tumor characteristics, such as tumor size or glycolysis, which provide very indirect information about the effect of precision treatments on the pharmacological targets of tumors. Several advanced imaging modalities allow for...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Cancers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-6694/15/6/1751 |
| _version_ | 1797612990077337600 |
|---|---|
| author | Nesrin Mansouri Daniel Balvay Omar Zenteno Caterina Facchin Thulaciga Yoganathan Thomas Viel Joaquin Lopez Herraiz Bertrand Tavitian Mailyn Pérez-Liva |
| author_facet | Nesrin Mansouri Daniel Balvay Omar Zenteno Caterina Facchin Thulaciga Yoganathan Thomas Viel Joaquin Lopez Herraiz Bertrand Tavitian Mailyn Pérez-Liva |
| author_sort | Nesrin Mansouri |
| collection | DOAJ |
| description | The standard assessment of response to cancer treatments is based on gross tumor characteristics, such as tumor size or glycolysis, which provide very indirect information about the effect of precision treatments on the pharmacological targets of tumors. Several advanced imaging modalities allow for the visualization of targeted tumor hallmarks. Descriptors extracted from these images can help establishing new classifications of precision treatment response. We propose a machine learning (ML) framework to analyze metabolic–anatomical–vascular imaging features from positron emission tomography, ultrafast Doppler, and computed tomography in a mouse model of paraganglioma undergoing anti-angiogenic treatment with sunitinib. Imaging features from the follow-up of sunitinib-treated (<i>n</i> = 8, imaged once-per-week/6-weeks) and sham-treated (<i>n</i> = 8, imaged once-per-week/3-weeks) mice groups were dimensionally reduced and analyzed with hierarchical clustering Analysis (HCA). The classes extracted from HCA were used with 10 ML classifiers to find a generalized tumor stage prediction model, which was validated with an independent dataset of sunitinib-treated mice. HCA provided three stages of treatment response that were validated using the best-performing ML classifier. The Gaussian naive Bayes classifier showed the best performance, with a training accuracy of 98.7 and an average area under curve of 100. Our results show that metabolic–anatomical–vascular markers allow defining treatment response trajectories that reflect the efficacy of an anti-angiogenic drug on the tumor target hallmark. |
| first_indexed | 2024-03-11T06:48:39Z |
| format | Article |
| id | doaj.art-6776c2e3c8a84028b73060064b2c39b9 |
| institution | Directory Open Access Journal |
| issn | 2072-6694 |
| language | English |
| last_indexed | 2024-03-11T06:48:39Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cancers |
| spelling | doaj.art-6776c2e3c8a84028b73060064b2c39b92023-11-17T10:06:42ZengMDPI AGCancers2072-66942023-03-01156175110.3390/cancers15061751Machine Learning of Multi-Modal Tumor Imaging Reveals Trajectories of Response to Precision TreatmentNesrin Mansouri0Daniel Balvay1Omar Zenteno2Caterina Facchin3Thulaciga Yoganathan4Thomas Viel5Joaquin Lopez Herraiz6Bertrand Tavitian7Mailyn Pérez-Liva8INSERM, PARCC, Université Paris Cité, F-75015 Paris, FranceINSERM, PARCC, Université Paris Cité, F-75015 Paris, FranceINSERM, PARCC, Université Paris Cité, F-75015 Paris, FranceINSERM, PARCC, Université Paris Cité, F-75015 Paris, FranceINSERM, PARCC, Université Paris Cité, F-75015 Paris, FranceINSERM, PARCC, Université Paris Cité, F-75015 Paris, FranceNuclear Physics Group and IPARCOS, Department of Structure of Matter, Thermal Physics and Electronics, CEI Moncloa, Universidad Complutense de Madrid, 28040 Madrid, SpainINSERM, PARCC, Université Paris Cité, F-75015 Paris, FranceINSERM, PARCC, Université Paris Cité, F-75015 Paris, FranceThe standard assessment of response to cancer treatments is based on gross tumor characteristics, such as tumor size or glycolysis, which provide very indirect information about the effect of precision treatments on the pharmacological targets of tumors. Several advanced imaging modalities allow for the visualization of targeted tumor hallmarks. Descriptors extracted from these images can help establishing new classifications of precision treatment response. We propose a machine learning (ML) framework to analyze metabolic–anatomical–vascular imaging features from positron emission tomography, ultrafast Doppler, and computed tomography in a mouse model of paraganglioma undergoing anti-angiogenic treatment with sunitinib. Imaging features from the follow-up of sunitinib-treated (<i>n</i> = 8, imaged once-per-week/6-weeks) and sham-treated (<i>n</i> = 8, imaged once-per-week/3-weeks) mice groups were dimensionally reduced and analyzed with hierarchical clustering Analysis (HCA). The classes extracted from HCA were used with 10 ML classifiers to find a generalized tumor stage prediction model, which was validated with an independent dataset of sunitinib-treated mice. HCA provided three stages of treatment response that were validated using the best-performing ML classifier. The Gaussian naive Bayes classifier showed the best performance, with a training accuracy of 98.7 and an average area under curve of 100. Our results show that metabolic–anatomical–vascular markers allow defining treatment response trajectories that reflect the efficacy of an anti-angiogenic drug on the tumor target hallmark.https://www.mdpi.com/2072-6694/15/6/1751multi-modal imagingparagangliomamachine learninghierarchical clusteringtreatment response |
| spellingShingle | Nesrin Mansouri Daniel Balvay Omar Zenteno Caterina Facchin Thulaciga Yoganathan Thomas Viel Joaquin Lopez Herraiz Bertrand Tavitian Mailyn Pérez-Liva Machine Learning of Multi-Modal Tumor Imaging Reveals Trajectories of Response to Precision Treatment Cancers multi-modal imaging paraganglioma machine learning hierarchical clustering treatment response |
| title | Machine Learning of Multi-Modal Tumor Imaging Reveals Trajectories of Response to Precision Treatment |
| title_full | Machine Learning of Multi-Modal Tumor Imaging Reveals Trajectories of Response to Precision Treatment |
| title_fullStr | Machine Learning of Multi-Modal Tumor Imaging Reveals Trajectories of Response to Precision Treatment |
| title_full_unstemmed | Machine Learning of Multi-Modal Tumor Imaging Reveals Trajectories of Response to Precision Treatment |
| title_short | Machine Learning of Multi-Modal Tumor Imaging Reveals Trajectories of Response to Precision Treatment |
| title_sort | machine learning of multi modal tumor imaging reveals trajectories of response to precision treatment |
| topic | multi-modal imaging paraganglioma machine learning hierarchical clustering treatment response |
| url | https://www.mdpi.com/2072-6694/15/6/1751 |
| work_keys_str_mv | AT nesrinmansouri machinelearningofmultimodaltumorimagingrevealstrajectoriesofresponsetoprecisiontreatment AT danielbalvay machinelearningofmultimodaltumorimagingrevealstrajectoriesofresponsetoprecisiontreatment AT omarzenteno machinelearningofmultimodaltumorimagingrevealstrajectoriesofresponsetoprecisiontreatment AT caterinafacchin machinelearningofmultimodaltumorimagingrevealstrajectoriesofresponsetoprecisiontreatment AT thulacigayoganathan machinelearningofmultimodaltumorimagingrevealstrajectoriesofresponsetoprecisiontreatment AT thomasviel machinelearningofmultimodaltumorimagingrevealstrajectoriesofresponsetoprecisiontreatment AT joaquinlopezherraiz machinelearningofmultimodaltumorimagingrevealstrajectoriesofresponsetoprecisiontreatment AT bertrandtavitian machinelearningofmultimodaltumorimagingrevealstrajectoriesofresponsetoprecisiontreatment AT mailynperezliva machinelearningofmultimodaltumorimagingrevealstrajectoriesofresponsetoprecisiontreatment |