Paper in ACII gets nominated for best paper award

Our latest paper on the automatic recognition of multiple affective states in virtual rehabilitation by exploiting the dependency relationships to appear in 8th International Conference on Affective Computing & Intelligent Interaction (ACII 2019) has been nominated for best paper award. This paper has been developed in collaboration with Prof. Nadia Berthouze from University College London.

Congratulations Joel! And that´s two this year.

Two new publications in Pervasive Health

We hit Pervasive Health with two new publications. One continues with our effort to incorporate affective computing to the virtual rehabilitation environment. In this case, it advances the methodological aspects supporting the AI engine proposing a late fusion over a Bayesian classifier to compensate for missing signals. The second proposes an innovative easy to interpret score for continuous and unobtrusive monitoring of therapy impact in the out-patient daily living, the index of limb usage balance (ILUB). This is accompanied by an smart platform that capitalizes on smartwatches and a smartphone to quantify the ILUB ubiquitously. Congratulations to Joel and Silvia as the main authors of these two papers.

Dr. Herrera-Vega joins the project to bring GT to the cloud

Dr. Javier Herrera-Vega has now joined the team and begins his job on the 1st of November. Javier has been a close collaborator for many years, and he has been at INAOE since long under several roles. However, it is not until now that he will be formally joining the GT team. 

Javier is an excellent and experienced programmer, and will focus on bringing GT to the cloud. Looking forward to the next step! Welcome!

New RECITES grant

A much welcomed new grant from the Red Colaborativa de Investigación Traslacional para el Envejecimiento Saludable (RECITES). We will be investigating the effect of virtual rehabilitation on elders surviving stroke during the subacute stage and their following at home. How cool is that? The grant will support our research with 430,000.00 MXM.

This research intends to carry out an observational study of the effect of a virtual therapy initiated during the subactue stage and followed after the patient has been discharged from hospital but still continues his rehablitation as an outpatient from home. This new research grant follows our recent interest in monitoring the stroke patient remotely using the benefits of cloud computing (see grant  SmartSDK H2020-ICT-2016-INT-723174).

We intend to show that an early intervention sustained on high motivation (subsequent to the virtual therapy) may boost dexterity without harm for the patient according to the new evidences that early treatment is safe and feasible. Further, we also intend to show that dexterity immprovements are higher with early intervention than with the current late intervention at the chronic stage.

New MSc students joined the GT project

Two new MSc students, Silvia and Gerardo, from the new MSc on Biomedical Sciences and Technologies at INAOE are joining the Gesture Therapy team. Silvia, with her clinical skills will be working on assessing the impact of a virtual rehabilitation therapy (yep! that’s GT!) in acute stage of stroke rehabilitation at home. This is a very interesting project that will produce very novel data in subacute stage!. Gerardo, with his engineering skills, will be contributing to the development of the long sought finger controller. Welcome both!

New publication on IEEE Transactions on Affective Computing

Congratulations to Joel for his journal paper on the use of affective computing in virtual rehabilitation which has been accepted for publication at IEEE Transactions on Affective Computing. The paper presents a novel way to harness the patient’s affective state in order to afford greater patient personalization.

Four states: anxiety, pain, engagement and tiredness (either physical or psychological), are associatively predicted from system inputs such as hand location and gripping strength. Contributions of the paper are; (i) a multiresolution classifier built from Semi-Naïve Bayesian classifiers, and (ii) establishing predictive relations for the considered states from the motor proxies capitalizing on the proposed classifier with recognition levels sufficient for exploitation. 3D hand locations and gripping strength streams were recorded from 5 post-stroke patients whilst undergoing motor rehabilitation therapy administered through virtual rehabilitation along 10 sessions over 4 weeks. Features from the streams characterized the motor dynamics, while spontaneous manifestations of the states were labelled from concomitant videos by experts for supervised classification. The new classifier was compared against baseline support vector machine (SVM) and random forest (RF) with all three exhibiting comparable performances. Inference of the aforementioned states departing from chosen motor surrogates appears feasible, expediting increased personalization of virtual motor neurorehabilitation therapies.

We are going to the Rehab Week! Two new publications accepted

We are going to the Rehab Week in London! Two new publications have been accepted at the 6th International Neurorehabilitation Symposium (INRS) and 15th IEEE Conference on Rehabilitation Robotics (ICORR 2017). In the INRS publication, Joel reports his advances on the affective computing applied to virtual rehabilitation. In the ICORR publication, Patrick describes the automation of the Fugl-Mayer motor dexterity evaluation. Both have been accepted for oral presentations.

Up again! Server down finally fixed… almost a year later

It seems that after almost a year the server is finally up again! In the meantime we have had several news in the Gesture Therapy. Prof Sucar has won the National Science Award from the Mexican government, Patrick has develop an automatic recognition system for the assessment for motor dexterity, Viviana has been studying the effect of abstraction in the presentation of the task in the virtual environment, Arlem has explored how the distribution of elements in the virtual environment affects the cognitive workload, Joel has made substantial progress in the detection of emotions to incorporate affective elements to our beloved GT, Elías has put a brand new tracker which is more robust and responsive than the previous, and the user interface and internal architecture is undergoing deep changes for a better integrations among other. We will be reporting on these with more detail in the following days.

We truly hope that our server problems are now a thing of the past.