000294874 001__ 294874
000294874 005__ 20241210182915.0
000294874 0247_ $$2doi$$a10.48550/ARXIV.2406.02534
000294874 037__ $$aDKFZ-2024-02584
000294874 1001_ $$0P:(DE-He78)d2bf7126723ea8f6005ba141ea3c3e2c$$aXiao, Shuhan$$b0$$eFirst author$$udkfz
000294874 245__ $$aEnhancing predictive imaging biomarker discovery through treatment effect analysis
000294874 260__ $$barXiv$$c2024
000294874 3367_ $$0PUB:(DE-HGF)25$$2PUB:(DE-HGF)$$aPreprint$$bpreprint$$mpreprint$$s1733825647_31422
000294874 3367_ $$2ORCID$$aWORKING_PAPER
000294874 3367_ $$028$$2EndNote$$aElectronic Article
000294874 3367_ $$2DRIVER$$apreprint
000294874 3367_ $$2BibTeX$$aARTICLE
000294874 3367_ $$2DataCite$$aOutput Types/Working Paper
000294874 520__ $$aIdentifying predictive covariates, which forecast individual treatment effectiveness, is crucial for decision-making across different disciplines such as personalized medicine. These covariates, referred to as biomarkers, are extracted from pre-treatment data, often within randomized controlled trials, and should be distinguished from prognostic biomarkers, which are independent of treatment assignment. Our study focuses on discovering predictive imaging biomarkers, specific image features, by leveraging pre-treatment images to uncover new causal relationships. Unlike labor-intensive approaches relying on handcrafted features prone to bias, we present a novel task of directly learning predictive features from images. We propose an evaluation protocol to assess a model's ability to identify predictive imaging biomarkers and differentiate them from purely prognostic ones by employing statistical testing and a comprehensive analysis of image feature attribution. We explore the suitability of deep learning models originally developed for estimating the conditional average treatment effect (CATE) for this task, which have been assessed primarily for their precision of CATE estimation while overlooking the evaluation of imaging biomarker discovery. Our proof-of-concept analysis demonstrates the feasibility and potential of our approach in discovering and validating predictive imaging biomarkers from synthetic outcomes and real-world image datasets. Our code is available at \url{https://github.com/MIC-DKFZ/predictive_image_biomarker_analysis}.
000294874 536__ $$0G:(DE-HGF)POF4-315$$a315 - Bildgebung und Radioonkologie (POF4-315)$$cPOF4-315$$fPOF IV$$x0
000294874 588__ $$aDataset connected to DataCite
000294874 650_7 $$2Other$$aImage and Video Processing (eess.IV)
000294874 650_7 $$2Other$$aArtificial Intelligence (cs.AI)
000294874 650_7 $$2Other$$aComputer Vision and Pattern Recognition (cs.CV)
000294874 650_7 $$2Other$$aMachine Learning (cs.LG)
000294874 650_7 $$2Other$$aFOS: Electrical engineering, electronic engineering, information engineering
000294874 650_7 $$2Other$$aFOS: Computer and information sciences
000294874 7001_ $$0P:(DE-He78)fee0b9e9b2afd400b7afbc6b083cfa4a$$aKlein, Lukas$$b1$$udkfz
000294874 7001_ $$0P:(DE-He78)ce7813ed6ec6ac6cc92e67e89a54ca10$$aPetersen, Jens$$b2
000294874 7001_ $$0P:(DE-He78)3da06896bf2a50a84d40c33c3b7a9b3e$$aVollmuth, Philipp$$b3$$udkfz
000294874 7001_ $$0P:(DE-HGF)0$$aJaeger, Paul F.$$b4
000294874 7001_ $$0P:(DE-He78)33c74005e1ce56f7025c4f6be15321b3$$aMaier-Hein, Klaus$$b5$$eLast author$$udkfz
000294874 773__ $$a10.48550/ARXIV.2406.02534
000294874 909CO $$ooai:inrepo02.dkfz.de:294874$$pVDB
000294874 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)d2bf7126723ea8f6005ba141ea3c3e2c$$aDeutsches Krebsforschungszentrum$$b0$$kDKFZ
000294874 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)fee0b9e9b2afd400b7afbc6b083cfa4a$$aDeutsches Krebsforschungszentrum$$b1$$kDKFZ
000294874 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)ce7813ed6ec6ac6cc92e67e89a54ca10$$aDeutsches Krebsforschungszentrum$$b2$$kDKFZ
000294874 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)3da06896bf2a50a84d40c33c3b7a9b3e$$aDeutsches Krebsforschungszentrum$$b3$$kDKFZ
000294874 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-HGF)0$$aDeutsches Krebsforschungszentrum$$b4$$kDKFZ
000294874 9101_ $$0I:(DE-588b)2036810-0$$6P:(DE-He78)33c74005e1ce56f7025c4f6be15321b3$$aDeutsches Krebsforschungszentrum$$b5$$kDKFZ
000294874 9131_ $$0G:(DE-HGF)POF4-315$$1G:(DE-HGF)POF4-310$$2G:(DE-HGF)POF4-300$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bGesundheit$$lKrebsforschung$$vBildgebung und Radioonkologie$$x0
000294874 9141_ $$y2024
000294874 9201_ $$0I:(DE-He78)E230-20160331$$kE230$$lE230 Medizinische Bildverarbeitung$$x0
000294874 9201_ $$0I:(DE-He78)E290-20160331$$kE290$$lNWG Interaktives maschinelles Lernen$$x1
000294874 980__ $$apreprint
000294874 980__ $$aVDB
000294874 980__ $$aI:(DE-He78)E230-20160331
000294874 980__ $$aI:(DE-He78)E290-20160331
000294874 980__ $$aUNRESTRICTED