Machine learning has recently made impressive advances in applications ranging from computer vision to natural language processing and has been extensively used in bioinformatics or medical image analysis. However, the machine learning paradigm that relies on data-driven connectionism cannot deal with complex medical problems perfectly. On the other hand, much knowledge of medical experience is based on symbolic representation and expert reasoning. The effective representation and embedding of medical expertise can assist the optimization of machine learning algorithms and effectively improve their performance, interpretability, and reliability. Therefore, effectively integrating disease knowledge and machine learning can provide a new scientific research perspective to explore more practical applications of machine learning, profound learning in individualized medicine.
With this Research Topic, we aim to define new advances in our understanding of knowledge-embedded machine learning to solve problems in individualized medicine using genomic approach or medical image analysis.
The topic would like to include but is not limited to:
• Development and application of machine learning methods in cancer genomics, primarily focused on developing new strategies in cancer risk evaluation.
• Development and application of machine learning methods in image analysis (including the radiologic medical image/histology slide), mainly focused on machine learning to develop new predictive models in diagnosis and risk evaluation.
• Development and application of machine learning methods in multi-omics analysis or combined analysis of genomic and image data, especially focused on new techniques or applications for better integration or interpretability.
• Novel tools in translational medicine, including novel software and pipelines in high-throughput sequencing data analysis, radiologic medical image, and histology slide analysis.
Scope consideration for manuscripts:
Manuscripts can be submitted to this Research Topic via three journals. We recommend those involving molecular knowledge or association between molecular and image data to be submitted to Frontiers in Molecular Biomedicine, those focusing on bioinformatics to be submitted to Frontiers in Genetics, and those focusing on medical image and digital pathology to be submitted to Frontiers in Physiology.
Machine learning has recently made impressive advances in applications ranging from computer vision to natural language processing and has been extensively used in bioinformatics or medical image analysis. However, the machine learning paradigm that relies on data-driven connectionism cannot deal with complex medical problems perfectly. On the other hand, much knowledge of medical experience is based on symbolic representation and expert reasoning. The effective representation and embedding of medical expertise can assist the optimization of machine learning algorithms and effectively improve their performance, interpretability, and reliability. Therefore, effectively integrating disease knowledge and machine learning can provide a new scientific research perspective to explore more practical applications of machine learning, profound learning in individualized medicine.
With this Research Topic, we aim to define new advances in our understanding of knowledge-embedded machine learning to solve problems in individualized medicine using genomic approach or medical image analysis.
The topic would like to include but is not limited to:
• Development and application of machine learning methods in cancer genomics, primarily focused on developing new strategies in cancer risk evaluation.
• Development and application of machine learning methods in image analysis (including the radiologic medical image/histology slide), mainly focused on machine learning to develop new predictive models in diagnosis and risk evaluation.
• Development and application of machine learning methods in multi-omics analysis or combined analysis of genomic and image data, especially focused on new techniques or applications for better integration or interpretability.
• Novel tools in translational medicine, including novel software and pipelines in high-throughput sequencing data analysis, radiologic medical image, and histology slide analysis.
Scope consideration for manuscripts:
Manuscripts can be submitted to this Research Topic via three journals. We recommend those involving molecular knowledge or association between molecular and image data to be submitted to Frontiers in Molecular Biomedicine, those focusing on bioinformatics to be submitted to Frontiers in Genetics, and those focusing on medical image and digital pathology to be submitted to Frontiers in Physiology.