Dec. 07, 2021
13:00
-14:00

Building 96, Room 001

​​​​​​

Speaker 01:  Or Symonitz​

Title: Predicting Dynamic Recovery Following Rehabilitation Treatment Using Graph-Theory Analysis of Resting State fMRI in Chronic Post-Stroke Subjects


Abstract:

Stroke is associated with damage to neural tissue and is the leading cause of long-term motor disability in adults. Dynamic balance impairments are one of the most debilitating outcomes of stroke, leading to increased falls and loss of mobility. The recovery of motor functions following stroke was shown to be affected by the initial brain impairment, passage of time, and dosage of rehabilitation treatment.  We demonstrate here that a novel approach including machine learning and graph analysis applied on the resting state fMRI scans of stroke patients taken before a dynamic balance rehabilitation treatment can predict their recovery. Furthermore, we report that global features, describing the brain graph as a whole, are more informative than local features, describing individual regions,  for predicting the success of a rehabilitation, reaching accuracy levels of 86% .

 Furthermore, graph embedding using NetLSD Wave exhibits 86% and 80% accuracy when predicting the community balance and mobility (CBM) scale before and after the treatment respectively. These graph-theory based neural markers may contribute to enhancing recovery through improving the selection of rehabilitative treatments.

Or Symonitz.jpg

Bio:

Or Symonitz Is a Soldier and a master student in the department of Software and Information Systems Engineering, at Ben-Gurion University of the Negev (BGU), Beer-Sheva, Israel. Studying under the superposition of Dr. Rami Puzis and Dr. Lior Shmuelof from the Brain and Action Lab​.

His research topics at prediction of balance rehabilitation of patients after a treatment among stroke patients.  


Speaker 02:  Eliad Shem Tov

Title: Interpretable Context-Aware Recommender Systems Utilizing Evolutionary Algorithms


Abstract:

A context-aware recommender system (CARS) utilizes users' context to provide personalized services. Contextual information can be derived from sensors in order to improve the accuracy of the recommendations. In our work, we focus on CARSs with high-dimensional contextual information that typically impacts the recommendation model, for example, by increasing the model's dimensionality and sparsity. Generating accurate recommendations is not enough to constitute a useful system from the user's perspective, since the use of some contextual information may cause problems, such as draining the user's battery, raising privacy concerns, and more. Previous studies suggested reducing the amount of contextual information utilized by using domain knowledge to select the most suitable information. This approach is only applicable when the set of contexts is small enough to handle and sufficient for preventing sparsity. Moreover, hand-crafted context information may not represent an optimal set of features for the recommendation process. Another approach is to compress the contextual information into a denser latent space, but this may limit the ability to explain the recommended items to the users or compromise their trust. In this work, we present a multi-step approach for selecting low-dimensional subsets of contextual information and incorporating them explicitly within CARSs. At the core of our approach is a novel feature selection algorithm based on genetic algorithms, which outperforms state-of-the-art dimensionality reduction CARS algorithms by improving recommendation accuracy and interpretability. Over the course of evolution, thousands of diverse feature subsets are generated; a deep context-aware model is produced for each feature subset, and the subsets are stacked together. The resulting stacked model is accurate and only uses interpretable, explicit features. Our approach includes a mechanism of tuning the different underlying algorithms that affect user concerns, such as privacy and battery consumption. We evaluated our approach on two high-dimensional context-aware datasets derived from smartphones. An empirical analysis of our results confirms that our proposed approach outperforms state-of-the-art CARS models while improving transparency and interpretability for the user. In addition to the empirical results, we present several use cases, examples and methodology of how researchers, domain experts and CARS modelers can tweak the feature selection algorithm to improve various user concerns and interpretability.

eliad shem tov.jpg 

Bio:


Eliad Shem Tov is a MSc student at the Software and Information Systems Engineering Department. He is a student in Meitar, studying under the supervision of Prof. Lior Rokach and Doctor Achiya Elyasaf. Eliad finished his BSc in Software Engineering last semester (2021). ​