Kooperativer Bibliotheksverbund

UID:

Umfang: 1 online resource (480 pages)

Ausgabe: First edition.

ISBN: 0-12-822001-5

Inhalt: Machine Learning and Artificial Intelligence in Radiation Oncology: A Guide for Clinicians is designed for the application of practical concepts in machine learning to clinical radiation oncology. It addresses the existing void in a resource to educate practicing clinicians about how machine learning can be used to improve clinical and patient-centered outcomes

Anmerkung: Front Cover -- MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE IN RADIATION ONCOLOGY -- MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE IN RADIATION ONCOLOGY -- Copyright -- Contents -- Contributors -- Foreword from the editors -- 1 - Fundamentals and overview -- 1 - Fundamentals of machine learning -- Artificial intelligence and machine learning -- Capturing and quantifying experience -- Learning from experience -- Forms of learning -- Prediction -- Clustering -- Classification -- Reduction -- Potential sources for error -- Poor quality data -- Data bias and imbalance -- Information-poor features -- Cognitive bias -- Violating the i.i.d assumption -- Nature of automated learning -- Training error -- From linear models to complex learning algorithms -- Types of machine learning -- Generative versus discriminative learning approaches -- Supervised learning -- Bayesian learning -- Neural network learning -- Support vector learning -- Tree learning -- Unsupervised learning -- Semi-supervised learning -- Summary -- References -- 2 - Artificial intelligence, machine learning, and bioethics in clinical medicine -- Introduction -- Ethics, AI/ML, and responsibility -- Principles of bioethics and AI/ML in clinical medicine and research -- Respect for persons -- Harms and benefits -- Justice and fairness -- Ethical partnerships with AI/ML: recommendations for clinicians -- Conclusion -- References -- 3 - Machine learning applications in cancer genomics -- Introduction -- Overview of genomic technologies -- Microarrays -- RNA-seq -- DNA-seq -- Epigenetic and regulatory sequencing -- Applications of genomics in oncology -- Tumor subtyping -- Driver mutation discovery -- Biomarkers of outcome in clinical practice -- Pharmacogenomics -- Common hurdles of machine learning in genomics -- Challenges in data acquisition -- Data sparsity -- Inter-tumor heterogeneity. , Intra-tumor heterogeneity -- Other common data issues -- Future directions -- References -- 4 - Radiomics: "unlocking the potential of medical images for precision radiation oncology" -- Introduction -- Radiomics workflow -- Data acquisition and preparation -- Tumor segmentation -- Feature extraction and selection -- Feature selection, modeling - the significance of machine learning -- Challenges of machine learning -- Potential pitfalls in the radiomics pipeline -- Data acquisition during imaging -- Segmentation -- Feature extraction - feature selection algorithms -- Model development -- Recommendations for the standardization of radiomics research -- Image biomarker standardization initiative -- FAIR guiding principles -- Radiomics and distributed learning -- A practical roadmap for radiomics research in radiation oncology -- Finding the clinical question and hypothesis -- Database selection-curation, image preprocessing and feature selection -- Machine learning method -- Training, validation, and benchmarking -- Stability -- Translation of radiomics into the clinic -- Black-box versus interpretability -- Deployment, implementation, and commissioning of radiomics models -- Operational excellence versus outcome prediction -- Conclusion -- References -- 5 - Deep learning for medical image segmentation -- Clinical need for automated image segmentation -- Rationale of using deep learning for medical image segmentation -- Typical deep learning framework -- Practical considerations for segmentation model learning -- Image pre-processing -- Image patch selection -- Data augmentation -- Model fusion and output uncertainty assessment -- Role of international competitions in medical image segmentation -- AI-based image segmentation and beyond -- Conclusion -- References -- 6 - Natural language processing in oncology -- What is natural language processing?. , Introduction -- History -- The main challenge -- Advantages of free text -- The disadvantage of free text -- Solutions -- NLP in oncology -- Use cases for NLP in oncology -- Diagnosis and staging information -- Clinical trial cohort selection -- Treatment planning -- Progression and outcomes -- The role of the clinician in developing an NLP system -- Project conception -- Labeling and annotation -- Feature selection -- System evaluation and error analysis -- General NLP tasks and challenges -- Core NLP tasks -- Shared tasks -- GLUE, SuperGLUE and generalizability -- Medical NLP tasks and challenges -- Medical language -- Clinical NLP tasks and topics -- Clinical concept extraction and normalisation -- Context extraction -- Template extraction -- NLP approaches -- Rule-based -- Traditional machine learning -- Neural networks and deep learning -- Word embeddings -- Convolutional and recurrent neural networks -- Transformers -- Language models and contextual embeddings -- Tools and resources -- Tools -- Terminologies and ontologies -- Clinical corpora -- Clinical implementation of an NLP system -- Evaluation metrics -- Data access and compute -- Labeling a dataset -- Building and validating a model -- Conclusions, limitations and future directions -- References -- 7 - Evaluating machine learning models: From development to clinical deployment -- Introduction -- Model development -- Model performance evaluation -- Real world impact assessment -- Model development -- Objective selection -- Model selection: Choosing best models during training -- Parameter versus hyperparameter selection -- Transfer learning and pre-trained parameters -- Bias-variance trade-off -- Cross validation -- Validation: Assessing for models on independent data -- Hold out test data -- External test data -- Development set -- Inconsistent terminology -- Feature selection. , Model performance evaluation -- Building a confusion matrix -- Decision surfaces -- Binary classification performance measures -- True class-specific performance measures -- Predicted classification-specific performance measures -- Case study: Screening for ovarian cancer using proteomics -- Case study: Predicting short term mortality -- Receiver-operating characteristic curve -- Precision recall curve -- Regression -- Residuals -- Loss function -- Mean squared error -- R2 -- Real world impact assessment -- AI models as clinical decision support systems -- Logistical issues in deployment -- Continued evaluation/regulation -- Bias, disparities, ethics -- References -- 2 - Research applications -- 8 - Germline genomics in radiotherapy -- Overview of germline genomic analyses -- Genome-wide association studies -- Big data -- Modern clinical genomics -- Artificial intelligence in clinical genomics -- Genome wide association studies (GWAS) -- GWAS overview -- GWAS workflow -- Radiogenomics -- Radiogenomics GWAS -- Machine learning and radiogenomics -- ML and GWAS -- Use of ML methodology in radiogenomics research -- Detection of epistasis using ML -- Approaches to increasing statistical power using ML -- Data driven methods -- Filtering: Pre-processing independent of model -- Wrapper and embedded feature selection -- Multi-step feature selection in radiogenomics -- Conclusions -- References -- 9 - Tumor genomics in radiotherapy -- Introduction -- Cancer genomics -- Gene expression -- Genetic variations -- Epigenetic -- Types of genomic data -- Genomics in radiotherapy -- Bioinformatics of tumor genomics -- Statistical methods -- Meta-analysis -- Linear regression -- Cox model -- Machine learning and deep learning methods -- Fully-connected neural network -- Convolutional neural network -- Example applications -- Oncology diagnosis examples. , Conventional machine learning signatures -- Prediction of tumor sites of origin based on random forestRF -learned genomic signatures (Penson et al., 2019) -- Deep learning signatures -- Diagnosis of 12 cancer types through deep learning-based genomic signatures (Sun et al., 2019) -- Predicting cancer driver genes based on somatic mutations using deep convolutional networks (Luo, Ding, et al., 2019) -- Radiotherapy examples -- Conventional signatures -- A genome-based model for adjusting radiotherapy dose (2017) (Scott et al., 2017) -- Development of a 24-gene predictor of response to postoperative radiotherapy in prostate cancer (Zhao et al., 2016) -- Prediction of outcomes in NSCLC by a multiobjective Bayesian network (Luo et al., 2018) -- Deep learning signatures -- Prediction of tumor control in NSCLC after radiotherapy by a composite network architecture (Cui, Luo, Hsin Tseng, et al., ... -- Combining machine learning handcrafted features with deep learning latent variables for the prediction of radiation pneumon ... -- Challenges and recommendations -- Conclusions -- References -- 10 - Radiotherapy outcome prediction with medical imaging -- Introduction -- Image biomarkers from pre-treatment imaging and RT outcome -- Section 1: Pre-treatment CT image biomarkers for the prediction of prognosis and tumor control -- Section 2: Pre-treatment PET image information for the prediction of prognosis and tumor control -- Section 3: Pre-treatment MR image biomarkers for the prediction of prognosis and tumor control -- Section 4: Pre-treatment image biomarkers for radiation-induced toxicities -- Delta image biomarkers associations with radiation dose and outcome -- Section 5: Image biomarker changes during and after treatment and their relationship to dose -- Section 6: Image biomarker change relationship with treatment outcome and toxicities. , Future directions and challenges in the clinical implementation and application of image biomarker prediction models in RT.

Weitere Ausg.: Print version: Rosenstein, Barry S. Machine Learning and Artificial Intelligence in Radiation Oncology San Diego : Elsevier Science & Technology,c2023 ISBN 9780128220009

Sprache: Englisch

UID:

Umfang: 1 online resource (480 pages)

Ausgabe: First edition.

ISBN: 0-12-822001-5

Inhalt: Machine Learning and Artificial Intelligence in Radiation Oncology: A Guide for Clinicians is designed for the application of practical concepts in machine learning to clinical radiation oncology. It addresses the existing void in a resource to educate practicing clinicians about how machine learning can be used to improve clinical and patient-centered outcomes

Anmerkung: Front Cover -- MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE IN RADIATION ONCOLOGY -- MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE IN RADIATION ONCOLOGY -- Copyright -- Contents -- Contributors -- Foreword from the editors -- 1 - Fundamentals and overview -- 1 - Fundamentals of machine learning -- Artificial intelligence and machine learning -- Capturing and quantifying experience -- Learning from experience -- Forms of learning -- Prediction -- Clustering -- Classification -- Reduction -- Potential sources for error -- Poor quality data -- Data bias and imbalance -- Information-poor features -- Cognitive bias -- Violating the i.i.d assumption -- Nature of automated learning -- Training error -- From linear models to complex learning algorithms -- Types of machine learning -- Generative versus discriminative learning approaches -- Supervised learning -- Bayesian learning -- Neural network learning -- Support vector learning -- Tree learning -- Unsupervised learning -- Semi-supervised learning -- Summary -- References -- 2 - Artificial intelligence, machine learning, and bioethics in clinical medicine -- Introduction -- Ethics, AI/ML, and responsibility -- Principles of bioethics and AI/ML in clinical medicine and research -- Respect for persons -- Harms and benefits -- Justice and fairness -- Ethical partnerships with AI/ML: recommendations for clinicians -- Conclusion -- References -- 3 - Machine learning applications in cancer genomics -- Introduction -- Overview of genomic technologies -- Microarrays -- RNA-seq -- DNA-seq -- Epigenetic and regulatory sequencing -- Applications of genomics in oncology -- Tumor subtyping -- Driver mutation discovery -- Biomarkers of outcome in clinical practice -- Pharmacogenomics -- Common hurdles of machine learning in genomics -- Challenges in data acquisition -- Data sparsity -- Inter-tumor heterogeneity. , Intra-tumor heterogeneity -- Other common data issues -- Future directions -- References -- 4 - Radiomics: "unlocking the potential of medical images for precision radiation oncology" -- Introduction -- Radiomics workflow -- Data acquisition and preparation -- Tumor segmentation -- Feature extraction and selection -- Feature selection, modeling - the significance of machine learning -- Challenges of machine learning -- Potential pitfalls in the radiomics pipeline -- Data acquisition during imaging -- Segmentation -- Feature extraction - feature selection algorithms -- Model development -- Recommendations for the standardization of radiomics research -- Image biomarker standardization initiative -- FAIR guiding principles -- Radiomics and distributed learning -- A practical roadmap for radiomics research in radiation oncology -- Finding the clinical question and hypothesis -- Database selection-curation, image preprocessing and feature selection -- Machine learning method -- Training, validation, and benchmarking -- Stability -- Translation of radiomics into the clinic -- Black-box versus interpretability -- Deployment, implementation, and commissioning of radiomics models -- Operational excellence versus outcome prediction -- Conclusion -- References -- 5 - Deep learning for medical image segmentation -- Clinical need for automated image segmentation -- Rationale of using deep learning for medical image segmentation -- Typical deep learning framework -- Practical considerations for segmentation model learning -- Image pre-processing -- Image patch selection -- Data augmentation -- Model fusion and output uncertainty assessment -- Role of international competitions in medical image segmentation -- AI-based image segmentation and beyond -- Conclusion -- References -- 6 - Natural language processing in oncology -- What is natural language processing?. , Introduction -- History -- The main challenge -- Advantages of free text -- The disadvantage of free text -- Solutions -- NLP in oncology -- Use cases for NLP in oncology -- Diagnosis and staging information -- Clinical trial cohort selection -- Treatment planning -- Progression and outcomes -- The role of the clinician in developing an NLP system -- Project conception -- Labeling and annotation -- Feature selection -- System evaluation and error analysis -- General NLP tasks and challenges -- Core NLP tasks -- Shared tasks -- GLUE, SuperGLUE and generalizability -- Medical NLP tasks and challenges -- Medical language -- Clinical NLP tasks and topics -- Clinical concept extraction and normalisation -- Context extraction -- Template extraction -- NLP approaches -- Rule-based -- Traditional machine learning -- Neural networks and deep learning -- Word embeddings -- Convolutional and recurrent neural networks -- Transformers -- Language models and contextual embeddings -- Tools and resources -- Tools -- Terminologies and ontologies -- Clinical corpora -- Clinical implementation of an NLP system -- Evaluation metrics -- Data access and compute -- Labeling a dataset -- Building and validating a model -- Conclusions, limitations and future directions -- References -- 7 - Evaluating machine learning models: From development to clinical deployment -- Introduction -- Model development -- Model performance evaluation -- Real world impact assessment -- Model development -- Objective selection -- Model selection: Choosing best models during training -- Parameter versus hyperparameter selection -- Transfer learning and pre-trained parameters -- Bias-variance trade-off -- Cross validation -- Validation: Assessing for models on independent data -- Hold out test data -- External test data -- Development set -- Inconsistent terminology -- Feature selection. , Model performance evaluation -- Building a confusion matrix -- Decision surfaces -- Binary classification performance measures -- True class-specific performance measures -- Predicted classification-specific performance measures -- Case study: Screening for ovarian cancer using proteomics -- Case study: Predicting short term mortality -- Receiver-operating characteristic curve -- Precision recall curve -- Regression -- Residuals -- Loss function -- Mean squared error -- R2 -- Real world impact assessment -- AI models as clinical decision support systems -- Logistical issues in deployment -- Continued evaluation/regulation -- Bias, disparities, ethics -- References -- 2 - Research applications -- 8 - Germline genomics in radiotherapy -- Overview of germline genomic analyses -- Genome-wide association studies -- Big data -- Modern clinical genomics -- Artificial intelligence in clinical genomics -- Genome wide association studies (GWAS) -- GWAS overview -- GWAS workflow -- Radiogenomics -- Radiogenomics GWAS -- Machine learning and radiogenomics -- ML and GWAS -- Use of ML methodology in radiogenomics research -- Detection of epistasis using ML -- Approaches to increasing statistical power using ML -- Data driven methods -- Filtering: Pre-processing independent of model -- Wrapper and embedded feature selection -- Multi-step feature selection in radiogenomics -- Conclusions -- References -- 9 - Tumor genomics in radiotherapy -- Introduction -- Cancer genomics -- Gene expression -- Genetic variations -- Epigenetic -- Types of genomic data -- Genomics in radiotherapy -- Bioinformatics of tumor genomics -- Statistical methods -- Meta-analysis -- Linear regression -- Cox model -- Machine learning and deep learning methods -- Fully-connected neural network -- Convolutional neural network -- Example applications -- Oncology diagnosis examples. , Conventional machine learning signatures -- Prediction of tumor sites of origin based on random forestRF -learned genomic signatures (Penson et al., 2019) -- Deep learning signatures -- Diagnosis of 12 cancer types through deep learning-based genomic signatures (Sun et al., 2019) -- Predicting cancer driver genes based on somatic mutations using deep convolutional networks (Luo, Ding, et al., 2019) -- Radiotherapy examples -- Conventional signatures -- A genome-based model for adjusting radiotherapy dose (2017) (Scott et al., 2017) -- Development of a 24-gene predictor of response to postoperative radiotherapy in prostate cancer (Zhao et al., 2016) -- Prediction of outcomes in NSCLC by a multiobjective Bayesian network (Luo et al., 2018) -- Deep learning signatures -- Prediction of tumor control in NSCLC after radiotherapy by a composite network architecture (Cui, Luo, Hsin Tseng, et al., ... -- Combining machine learning handcrafted features with deep learning latent variables for the prediction of radiation pneumon ... -- Challenges and recommendations -- Conclusions -- References -- 10 - Radiotherapy outcome prediction with medical imaging -- Introduction -- Image biomarkers from pre-treatment imaging and RT outcome -- Section 1: Pre-treatment CT image biomarkers for the prediction of prognosis and tumor control -- Section 2: Pre-treatment PET image information for the prediction of prognosis and tumor control -- Section 3: Pre-treatment MR image biomarkers for the prediction of prognosis and tumor control -- Section 4: Pre-treatment image biomarkers for radiation-induced toxicities -- Delta image biomarkers associations with radiation dose and outcome -- Section 5: Image biomarker changes during and after treatment and their relationship to dose -- Section 6: Image biomarker change relationship with treatment outcome and toxicities. , Future directions and challenges in the clinical implementation and application of image biomarker prediction models in RT.

Weitere Ausg.: Print version: Rosenstein, Barry S. Machine Learning and Artificial Intelligence in Radiation Oncology San Diego : Elsevier Science & Technology,c2023 ISBN 9780128220009

Sprache: Englisch

UID:

Umfang: 1 online resource (480 pages)

Ausgabe: First edition.

ISBN: 0-12-822001-5

Inhalt: Machine Learning and Artificial Intelligence in Radiation Oncology: A Guide for Clinicians is designed for the application of practical concepts in machine learning to clinical radiation oncology. It addresses the existing void in a resource to educate practicing clinicians about how machine learning can be used to improve clinical and patient-centered outcomes

Anmerkung: Front Cover -- MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE IN RADIATION ONCOLOGY -- MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE IN RADIATION ONCOLOGY -- Copyright -- Contents -- Contributors -- Foreword from the editors -- 1 - Fundamentals and overview -- 1 - Fundamentals of machine learning -- Artificial intelligence and machine learning -- Capturing and quantifying experience -- Learning from experience -- Forms of learning -- Prediction -- Clustering -- Classification -- Reduction -- Potential sources for error -- Poor quality data -- Data bias and imbalance -- Information-poor features -- Cognitive bias -- Violating the i.i.d assumption -- Nature of automated learning -- Training error -- From linear models to complex learning algorithms -- Types of machine learning -- Generative versus discriminative learning approaches -- Supervised learning -- Bayesian learning -- Neural network learning -- Support vector learning -- Tree learning -- Unsupervised learning -- Semi-supervised learning -- Summary -- References -- 2 - Artificial intelligence, machine learning, and bioethics in clinical medicine -- Introduction -- Ethics, AI/ML, and responsibility -- Principles of bioethics and AI/ML in clinical medicine and research -- Respect for persons -- Harms and benefits -- Justice and fairness -- Ethical partnerships with AI/ML: recommendations for clinicians -- Conclusion -- References -- 3 - Machine learning applications in cancer genomics -- Introduction -- Overview of genomic technologies -- Microarrays -- RNA-seq -- DNA-seq -- Epigenetic and regulatory sequencing -- Applications of genomics in oncology -- Tumor subtyping -- Driver mutation discovery -- Biomarkers of outcome in clinical practice -- Pharmacogenomics -- Common hurdles of machine learning in genomics -- Challenges in data acquisition -- Data sparsity -- Inter-tumor heterogeneity. , Intra-tumor heterogeneity -- Other common data issues -- Future directions -- References -- 4 - Radiomics: "unlocking the potential of medical images for precision radiation oncology" -- Introduction -- Radiomics workflow -- Data acquisition and preparation -- Tumor segmentation -- Feature extraction and selection -- Feature selection, modeling - the significance of machine learning -- Challenges of machine learning -- Potential pitfalls in the radiomics pipeline -- Data acquisition during imaging -- Segmentation -- Feature extraction - feature selection algorithms -- Model development -- Recommendations for the standardization of radiomics research -- Image biomarker standardization initiative -- FAIR guiding principles -- Radiomics and distributed learning -- A practical roadmap for radiomics research in radiation oncology -- Finding the clinical question and hypothesis -- Database selection-curation, image preprocessing and feature selection -- Machine learning method -- Training, validation, and benchmarking -- Stability -- Translation of radiomics into the clinic -- Black-box versus interpretability -- Deployment, implementation, and commissioning of radiomics models -- Operational excellence versus outcome prediction -- Conclusion -- References -- 5 - Deep learning for medical image segmentation -- Clinical need for automated image segmentation -- Rationale of using deep learning for medical image segmentation -- Typical deep learning framework -- Practical considerations for segmentation model learning -- Image pre-processing -- Image patch selection -- Data augmentation -- Model fusion and output uncertainty assessment -- Role of international competitions in medical image segmentation -- AI-based image segmentation and beyond -- Conclusion -- References -- 6 - Natural language processing in oncology -- What is natural language processing?. , Introduction -- History -- The main challenge -- Advantages of free text -- The disadvantage of free text -- Solutions -- NLP in oncology -- Use cases for NLP in oncology -- Diagnosis and staging information -- Clinical trial cohort selection -- Treatment planning -- Progression and outcomes -- The role of the clinician in developing an NLP system -- Project conception -- Labeling and annotation -- Feature selection -- System evaluation and error analysis -- General NLP tasks and challenges -- Core NLP tasks -- Shared tasks -- GLUE, SuperGLUE and generalizability -- Medical NLP tasks and challenges -- Medical language -- Clinical NLP tasks and topics -- Clinical concept extraction and normalisation -- Context extraction -- Template extraction -- NLP approaches -- Rule-based -- Traditional machine learning -- Neural networks and deep learning -- Word embeddings -- Convolutional and recurrent neural networks -- Transformers -- Language models and contextual embeddings -- Tools and resources -- Tools -- Terminologies and ontologies -- Clinical corpora -- Clinical implementation of an NLP system -- Evaluation metrics -- Data access and compute -- Labeling a dataset -- Building and validating a model -- Conclusions, limitations and future directions -- References -- 7 - Evaluating machine learning models: From development to clinical deployment -- Introduction -- Model development -- Model performance evaluation -- Real world impact assessment -- Model development -- Objective selection -- Model selection: Choosing best models during training -- Parameter versus hyperparameter selection -- Transfer learning and pre-trained parameters -- Bias-variance trade-off -- Cross validation -- Validation: Assessing for models on independent data -- Hold out test data -- External test data -- Development set -- Inconsistent terminology -- Feature selection. , Model performance evaluation -- Building a confusion matrix -- Decision surfaces -- Binary classification performance measures -- True class-specific performance measures -- Predicted classification-specific performance measures -- Case study: Screening for ovarian cancer using proteomics -- Case study: Predicting short term mortality -- Receiver-operating characteristic curve -- Precision recall curve -- Regression -- Residuals -- Loss function -- Mean squared error -- R2 -- Real world impact assessment -- AI models as clinical decision support systems -- Logistical issues in deployment -- Continued evaluation/regulation -- Bias, disparities, ethics -- References -- 2 - Research applications -- 8 - Germline genomics in radiotherapy -- Overview of germline genomic analyses -- Genome-wide association studies -- Big data -- Modern clinical genomics -- Artificial intelligence in clinical genomics -- Genome wide association studies (GWAS) -- GWAS overview -- GWAS workflow -- Radiogenomics -- Radiogenomics GWAS -- Machine learning and radiogenomics -- ML and GWAS -- Use of ML methodology in radiogenomics research -- Detection of epistasis using ML -- Approaches to increasing statistical power using ML -- Data driven methods -- Filtering: Pre-processing independent of model -- Wrapper and embedded feature selection -- Multi-step feature selection in radiogenomics -- Conclusions -- References -- 9 - Tumor genomics in radiotherapy -- Introduction -- Cancer genomics -- Gene expression -- Genetic variations -- Epigenetic -- Types of genomic data -- Genomics in radiotherapy -- Bioinformatics of tumor genomics -- Statistical methods -- Meta-analysis -- Linear regression -- Cox model -- Machine learning and deep learning methods -- Fully-connected neural network -- Convolutional neural network -- Example applications -- Oncology diagnosis examples. , Conventional machine learning signatures -- Prediction of tumor sites of origin based on random forestRF -learned genomic signatures (Penson et al., 2019) -- Deep learning signatures -- Diagnosis of 12 cancer types through deep learning-based genomic signatures (Sun et al., 2019) -- Predicting cancer driver genes based on somatic mutations using deep convolutional networks (Luo, Ding, et al., 2019) -- Radiotherapy examples -- Conventional signatures -- A genome-based model for adjusting radiotherapy dose (2017) (Scott et al., 2017) -- Development of a 24-gene predictor of response to postoperative radiotherapy in prostate cancer (Zhao et al., 2016) -- Prediction of outcomes in NSCLC by a multiobjective Bayesian network (Luo et al., 2018) -- Deep learning signatures -- Prediction of tumor control in NSCLC after radiotherapy by a composite network architecture (Cui, Luo, Hsin Tseng, et al., ... -- Combining machine learning handcrafted features with deep learning latent variables for the prediction of radiation pneumon ... -- Challenges and recommendations -- Conclusions -- References -- 10 - Radiotherapy outcome prediction with medical imaging -- Introduction -- Image biomarkers from pre-treatment imaging and RT outcome -- Section 1: Pre-treatment CT image biomarkers for the prediction of prognosis and tumor control -- Section 2: Pre-treatment PET image information for the prediction of prognosis and tumor control -- Section 3: Pre-treatment MR image biomarkers for the prediction of prognosis and tumor control -- Section 4: Pre-treatment image biomarkers for radiation-induced toxicities -- Delta image biomarkers associations with radiation dose and outcome -- Section 5: Image biomarker changes during and after treatment and their relationship to dose -- Section 6: Image biomarker change relationship with treatment outcome and toxicities. , Future directions and challenges in the clinical implementation and application of image biomarker prediction models in RT.

Weitere Ausg.: Print version: Rosenstein, Barry S. Machine Learning and Artificial Intelligence in Radiation Oncology San Diego : Elsevier Science & Technology,c2023 ISBN 9780128220009

Sprache: Englisch