{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/156772"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/156772","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"Predicting Patient Outcomes in the EPOCH Clinical Trial","abstract":"Metastatic colorectal cancer (mCRC) has a poor prognosis and high mortality rate, but innovative therapies such as transarterial radioembolization (TARE) can improve patient outcomes. The EPOCHclinical trial demonstrated that TARE improved hepatic progressionfree survival (hPFS) in patients with colorectal liver metastases, and computational methods to analyze the multimodal data collected can identify patient subgroups and predict treatment response for personalized medicine. First, a comprehensive data preprocessing pipeline curated a high-quality dataset of liver-region Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) scans paired with patient biomarkers. Multi-Dimensional Subset Scanning (MDSS) identified a group of patients with shared biomarkers that exhibited poor response to TARE, and Cox Proportional Hazards (CoxPH) modeling revealed hazard ratios for biomarkers aligning with clinical expectations, albeit with a limited C-index. Augmenting CoxPH modeling with embeddings from a deep learning foundation model pre-trained on liver CT and MRI scans and fine-tuned to predict treatment response resulted in a substantially higher C-index. Interestingly, models fine-tuned to predict one clinical feature had improved predictive accuracy for other features they were not specifically trained on, and Class Activation Mapping (CAM) visualizations showed that salient embedding dimensions focus on the liver region, providing interpretability. The ensemble of computational techniques applied to multimodal clinical trial data successfully identified patient subgroups, extracted predictive biomarkers, and enhanced the accuracy of treatment response predictions, contributing to the development of more effective, personalized treatment strategies for mCRC patients undergoing TARE.","abstract_html":"Metastatic colorectal cancer (mCRC) has a poor prognosis and high mortality rate, but innovative therapies such as transarterial radioembolization (TARE) can improve patient outcomes. The EPOCHclinical trial demonstrated that TARE improved hepatic progressionfree survival (hPFS) in patients with colorectal liver metastases, and computational methods to analyze the multimodal data collected can identify patient subgroups and predict treatment response for personalized medicine. First, a comprehensive data preprocessing pipeline curated a high-quality dataset of liver-region Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) scans paired with patient biomarkers. Multi-Dimensional Subset Scanning (MDSS) identified a group of patients with shared biomarkers that exhibited poor response to TARE, and Cox Proportional Hazards (CoxPH) modeling revealed hazard ratios for biomarkers aligning with clinical expectations, albeit with a limited C-index. Augmenting CoxPH modeling with embeddings from a deep learning foundation model pre-trained on liver CT and MRI scans and fine-tuned to predict treatment response resulted in a substantially higher C-index. Interestingly, models fine-tuned to predict one clinical feature had improved predictive accuracy for other features they were not specifically trained on, and Class Activation Mapping (CAM) visualizations showed that salient embedding dimensions focus on the liver region, providing interpretability. The ensemble of computational techniques applied to multimodal clinical trial data successfully identified patient subgroups, extracted predictive biomarkers, and enhanced the accuracy of treatment response predictions, contributing to the development of more effective, personalized treatment strategies for mCRC patients undergoing TARE.","abstract_has_math":false,"creators":["Parsan, Nithin"],"institution":"Massachusetts Institute of Technology","degree_name":"Master","degree_level":null,"degree_discipline":null,"degree_department":"Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","school":null,"contributors":[],"advisors":["Ng, Kenney","Oliva, Aude"],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-05","date_published":"2024-05","updated_at":"2026-07-22T22:21:55Z","subjects":[],"languages":[],"rights":["Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","Copyright retained by author(s)"],"rights_urls":["https://creativecommons.org/licenses/by-nc-nd/4.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/156772","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Ng, Kenney","Oliva, Aude"]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"]},{"key":"dc:creator","label":"Author","values":["Parsan, Nithin"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2024-09-16T13:48:12Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2024-09-16T13:48:12Z"]},{"key":"dc:date.issued","label":"Date","values":["2024-05"]},{"key":"dc:publisher","label":"Institution","values":["Massachusetts Institute of Technology"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master","Master of Engineering in Computer Science and Molecular Biology"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","Copyright retained by author(s)"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://creativecommons.org/licenses/by-nc-nd/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/156772"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Metastatic colorectal cancer (mCRC) has a poor prognosis and high mortality rate, but innovative therapies such as transarterial radioembolization (TARE) can improve patient outcomes. The EPOCHclinical trial demonstrated that TARE improved hepatic progressionfree survival (hPFS) in patients with colorectal liver metastases, and computational methods to analyze the multimodal data collected can identify patient subgroups and predict treatment response for personalized medicine. First, a comprehensive data preprocessing pipeline curated a high-quality dataset of liver-region Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) scans paired with patient biomarkers. Multi-Dimensional Subset Scanning (MDSS) identified a group of patients with shared biomarkers that exhibited poor response to TARE, and Cox Proportional Hazards (CoxPH) modeling revealed hazard ratios for biomarkers aligning with clinical expectations, albeit with a limited C-index. Augmenting CoxPH modeling with embeddings from a deep learning foundation model pre-trained on liver CT and MRI scans and fine-tuned to predict treatment response resulted in a substantially higher C-index. Interestingly, models fine-tuned to predict one clinical feature had improved predictive accuracy for other features they were not specifically trained on, and Class Activation Mapping (CAM) visualizations showed that salient embedding dimensions focus on the liver region, providing interpretability. The ensemble of computational techniques applied to multimodal clinical trial data successfully identified patient subgroups, extracted predictive biomarkers, and enhanced the accuracy of treatment response predictions, contributing to the development of more effective, personalized treatment strategies for mCRC patients undergoing TARE."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["M.Eng."]},{"key":"dc:title","label":"Title","values":["Predicting Patient Outcomes in the EPOCH Clinical Trial"]}]}],"canonical_facts":{"dc:contributor.advisor":["Ng, Kenney","Oliva, Aude"],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"],"dc:creator":["Parsan, Nithin"],"dc:date.accessioned":["2024-09-16T13:48:12Z"],"dc:date.available":["2024-09-16T13:48:12Z"],"dc:date.issued":["2024-05"],"dc:description.abstract":["Metastatic colorectal cancer (mCRC) has a poor prognosis and high mortality rate, but innovative therapies such as transarterial radioembolization (TARE) can improve patient outcomes. The EPOCHclinical trial demonstrated that TARE improved hepatic progressionfree survival (hPFS) in patients with colorectal liver metastases, and computational methods to analyze the multimodal data collected can identify patient subgroups and predict treatment response for personalized medicine. First, a comprehensive data preprocessing pipeline curated a high-quality dataset of liver-region Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) scans paired with patient biomarkers. Multi-Dimensional Subset Scanning (MDSS) identified a group of patients with shared biomarkers that exhibited poor response to TARE, and Cox Proportional Hazards (CoxPH) modeling revealed hazard ratios for biomarkers aligning with clinical expectations, albeit with a limited C-index. Augmenting CoxPH modeling with embeddings from a deep learning foundation model pre-trained on liver CT and MRI scans and fine-tuned to predict treatment response resulted in a substantially higher C-index. Interestingly, models fine-tuned to predict one clinical feature had improved predictive accuracy for other features they were not specifically trained on, and Class Activation Mapping (CAM) visualizations showed that salient embedding dimensions focus on the liver region, providing interpretability. The ensemble of computational techniques applied to multimodal clinical trial data successfully identified patient subgroups, extracted predictive biomarkers, and enhanced the accuracy of treatment response predictions, contributing to the development of more effective, personalized treatment strategies for mCRC patients undergoing TARE."],"dc:description.degree":["M.Eng."],"dc:identifier.uri":["https://hdl.handle.net/1721.1/156772"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","Copyright retained by author(s)"],"dc:rights.uri":["https://creativecommons.org/licenses/by-nc-nd/4.0/"],"dc:title":["Predicting Patient Outcomes in the EPOCH Clinical Trial"],"dc:type":["Thesis"],"thesis:degree_name":["Master","Master of Engineering in Computer Science and Molecular Biology"]},"updated_at":"2026-07-22T22:21:55Z"}