Cardiovascular disease has been a leading death cause of human beings among cancer survivors (almost 370,000 each year around the world), followed by the death cause of new tumor development or cancer recurrence. In recent years, cardio-oncology arises as the growing awareness of side cardiac dysfunctions from cancer therapies such as chemotherapy, radiotherapy, immunotherapy, etc. Early detection, monitoring, prevention, and treatment of cardiovascular toxicity in cancer therapies is increasingly becoming urgent in precision cardio-oncology preventive care.
Precision cardio-oncology emphasizes individualized algorithms based on cardiovascular risk, cancer, and cancer treatments, and it considers treatment risks from oncologic care. For instance, patient-specific tumor genetics and therapy-specific cardiovascular risks are useful to redefine tumor classification and dictate cancer treatment algorithms. Artificial intelligence (AI)-based predictive analytics (supervised learning and unsupervised learning) are just available to use in recent decade in capturing dependent patterns between cardiovascular diseases and cancer therapies, which provides transformative analytical routes for prediction and diagnosis of individualized cardio-oncology risks. Notably, interpretable machine learning models show promises from its strengths in accurately identifying and capturing patient-specific complex hidden interaction patterns of longitudinal/multimodal data (omics, ECGs, drugs/therapies) with cardiac dysfunctions for clinical cancer treatment decisions in a methodologically explainable way.
This Research Topic aims to create a forum for current advances of interpretable predictive analytics in AI development to deal with the challenges regarding precision and accuracy of cardiotoxicity risk assessment in cancer survivors. Of special interest are: (I) innovative methodologies that ease the interpretability of AI predictions towards the validation of respective diagnostic, prognostic, and therapeutic models for side cardiac dysfunctions/complications before, during, or after cancer therapies; and (II) translational medical implications of interpretable AI-based predictive risk analytics for cancer therapy-related cardiac dysfunctions, including novel automatic data processing techniques for massive growth of longitudinal/multimodal heterogeneous data on cardio-oncology, development and configuration of advanced cardiotoxicity workflows or systems that tailored innovative data patterns discovery/visualization, risk predictions, and treatment recommendations in cardio-oncology care.
This Research Topic reports latest interdisciplinary research on developing novel interpretable predictive analytics for precision cardio-oncology preventive care, to capitalize on cardio-oncology big data.
- Interpretable machine learning/deep learning techniques in cardio-oncology complications treatment
- Institutional cardio-oncology data to predict adverse cardiac outcomes
- ECG/Image data for prediction and prognostication in cardio-oncology
- Biologically relevant AI models in precision and translational cardio-oncology
- Novel data patterns discovery/visualization in cardiology clinical practice
- Data science and AI techniques to discover new cardiotoxic pharmacologic agents
- Network analytics for longitudinal/multimodal heterogeneous cardio-oncology data
- Dynamic complexity of evolving cardio-oncology processes
- Predictive data analytics in chemotherapy, radiotherapy, and immunotherapy-related cardiotoxicity
- Interpretable bioinformatics for drug discovery in cardio-oncology
- Integrative systems for predictive cardiology treatment recommendation
- Translational real applications of predictive analytics for cardiovascular toxicities preventive care
Cardiovascular disease has been a leading death cause of human beings among cancer survivors (almost 370,000 each year around the world), followed by the death cause of new tumor development or cancer recurrence. In recent years, cardio-oncology arises as the growing awareness of side cardiac dysfunctions from cancer therapies such as chemotherapy, radiotherapy, immunotherapy, etc. Early detection, monitoring, prevention, and treatment of cardiovascular toxicity in cancer therapies is increasingly becoming urgent in precision cardio-oncology preventive care.
Precision cardio-oncology emphasizes individualized algorithms based on cardiovascular risk, cancer, and cancer treatments, and it considers treatment risks from oncologic care. For instance, patient-specific tumor genetics and therapy-specific cardiovascular risks are useful to redefine tumor classification and dictate cancer treatment algorithms. Artificial intelligence (AI)-based predictive analytics (supervised learning and unsupervised learning) are just available to use in recent decade in capturing dependent patterns between cardiovascular diseases and cancer therapies, which provides transformative analytical routes for prediction and diagnosis of individualized cardio-oncology risks. Notably, interpretable machine learning models show promises from its strengths in accurately identifying and capturing patient-specific complex hidden interaction patterns of longitudinal/multimodal data (omics, ECGs, drugs/therapies) with cardiac dysfunctions for clinical cancer treatment decisions in a methodologically explainable way.
This Research Topic aims to create a forum for current advances of interpretable predictive analytics in AI development to deal with the challenges regarding precision and accuracy of cardiotoxicity risk assessment in cancer survivors. Of special interest are: (I) innovative methodologies that ease the interpretability of AI predictions towards the validation of respective diagnostic, prognostic, and therapeutic models for side cardiac dysfunctions/complications before, during, or after cancer therapies; and (II) translational medical implications of interpretable AI-based predictive risk analytics for cancer therapy-related cardiac dysfunctions, including novel automatic data processing techniques for massive growth of longitudinal/multimodal heterogeneous data on cardio-oncology, development and configuration of advanced cardiotoxicity workflows or systems that tailored innovative data patterns discovery/visualization, risk predictions, and treatment recommendations in cardio-oncology care.
This Research Topic reports latest interdisciplinary research on developing novel interpretable predictive analytics for precision cardio-oncology preventive care, to capitalize on cardio-oncology big data.
- Interpretable machine learning/deep learning techniques in cardio-oncology complications treatment
- Institutional cardio-oncology data to predict adverse cardiac outcomes
- ECG/Image data for prediction and prognostication in cardio-oncology
- Biologically relevant AI models in precision and translational cardio-oncology
- Novel data patterns discovery/visualization in cardiology clinical practice
- Data science and AI techniques to discover new cardiotoxic pharmacologic agents
- Network analytics for longitudinal/multimodal heterogeneous cardio-oncology data
- Dynamic complexity of evolving cardio-oncology processes
- Predictive data analytics in chemotherapy, radiotherapy, and immunotherapy-related cardiotoxicity
- Interpretable bioinformatics for drug discovery in cardio-oncology
- Integrative systems for predictive cardiology treatment recommendation
- Translational real applications of predictive analytics for cardiovascular toxicities preventive care