Quantifying differences in feature importance rankings of #machinelearning #classification could enhance #interpretability and #explainability: we show how through the rank-biased overlap similarity measure. Take a look at my novel work!
Ever wondered how to quantitatively compare feature importance produced by Machine Learning algorithms?
In this new work presented at the Brain Informatics 2022, we introduce the Rank-Biased Overlap (RBO) as similarity measure for comparing rankings of features ordered by their importance. We used the automatic classification of Parkinson’s disease as case study.
Take a look at my recording if you are curious!
In case you missed the live, here the recording of the BI2022 Special Session on EXPLAINABLE ARTIFICIAL INTELLIGENCE FOR UNVEILING THE BRAIN: FROM THE BLACK-BOX TO THE GLASS-BOX (XAIB)
15th July 2022, 14:00-16:00 (GMT+2), with Prof. Monica Hernandez, Dr. Bojan Bogdanovic and Dr. Antonio Parziale
Frontiers in Neurology – Applied Neuroimaging
Brain Hemispheric Specialization and its pathological change revealed by Neuroimaging and Neuropsychology
Brain hemispheric specialization has a fundamental role in the functional cortical organization in humans and the morphological, functional, metabolic and connectivity asymmetries of brain regions were traditionally correlated to the optimal information processing, language function, visuospatial task, attention, and many aspects of emotion.
The laterality of human brain varies with the aging and the variations from the normal pattern of asymmetry could be often suggestive of pathology. Indeed, it has been suggested that abnormal asymmetry may serve as a neuroanatomical marker or as a risk factor. In other words, the existence of asymmetry in brain regions where the symmetry is expected or, on the contrary, the absence of asymmetry where asymmetry is expected could be often indicative of neurological or neurodegenerative disorder.
Our aim is to investigate whether and how the brain asymmetries change, and the abnormalities could cause behavioral, cognitive, and symptomatic alterations in neurological diseases, explored through the new advanced technology in the field.
The present Research Topic is to collect scientific works on the brain lateralization investigated with Neuroimaging approaches and Neuropsychological assessments in neurological disease. The following contributions are welcome:
• New research on biomarkers of brain lateralization;
• Novel neuroimaging methods to characterize the brain asymmetry;
• Novel neuropsychological assessments and techniques to better identify the brain specialization;
• Correlation study between neuroimaging findings and neuropsychological evaluations;
• Works exploring clinical conditions of asymmetry through multi-modal imaging approaches such as MRI, fMRI, DTI, PET or MEG;
• Neuroimaging and Neuropsychological asymmetrical data investigations through Artificial Intelligence and Machine Learning;
• Sex- and age-linked differences in brain asymmetry and neuropsychology patterns;
• Longitudinal studies on the brain asymmetry progression in normal aging or diseases;
• Identification of new metrics or indices for calculating the brain asymmetry;
• Review articles that summarize the current literature on the brain hemispheres specialization.
My Editorial for the Special Issue on “Machine Learning in Healthcare and Biomedical Application”
++++ CALL FOR PAPERS ++++
The wide application of artificial intelligence (AI) and machine learning (ML) on neuroscience data represents an unprecedented way for understanding neurological diseases. The AI and ML could, indeed, deal with multi-modal, multi-dimensional, and multi-source data, which can help to extract new knowledge about the pathological mechanisms that affect the human brain and, more generally, the nervous system. This Special Issue of the Journal of Personalized Medicine is devoted to collect original scientific articles that explore neurological diseases through the use of AI and ML approaches. In particular, we accept works that apply supervised and unsupervised learning, reinforcement learning, deep learning, and the more recent explainable and interpretable ML methodologies. Moreover, we seek to collect studies using and exploring different source of data, such as neuroimaging (structural MRI, functional MRI and Nirs), neurophysiology (TMS, EMG, EEG, MEG), biorobotics, and biomechanics (inertial, wearable, IoT sensors) applied to neurodegenerative diseases.
Contributions such as systematic reviews or meta-analyses on the above-mentioned topics are also welcome.
Dr. Alessia Sarica
Dr. Vera Gramigna
Dr. Maria Giovanna Bianco
In case you missed the live, here the recording of the BI2021 Special Session on EXPLAINABLE ARTIFICIAL INTELLIGENCE FOR UNVEILING THE BRAIN: FROM THE BLACK-BOX TO THE GLASS-BOX (XAIB)
18th September 2021, 14:00-16:00 UK time (GMT+1), with Dr. Rich Caruana, Dr. Michele Ferrante and Dr. Dimitris Pinotsis:
A great virtual OHBM this year! I proposed the application of Interpretable Artificial Intelligence on Neuroimaging data through the Explainable Boosting Machine on Alzheimer’s data. Here my poster and a video presenting my work.