Milena Korostenskaja, PhD
Neuroscientist, Certified Life Coach

Dr. Milena Korostenskaja (Dr. K) is a Founder of The Institute of Neuroapproaches. She offers STEP OUT OF BURNOUT life coaching program to to help her peers in science and academia overcome burnout.


"Advancing the field of pre-surgical functional brain mapping: a multidisciplinary approach with incorporated machine learning” invited talk at the Lithuanian Department of Statistics, Lithuanian Association of Statistics, Vilnius, Lithuania

November 26, 2019

Adaptive neurotechnologies enable real-time interactions with the nervous system. The primary goals of adaptive neurotechnologies include facilitation of changes in neuroplasticity; enhancement, restoration or replacement of neuromuscular functions; evaluation, and localization of brain activity. During Dr. Korostenskaja's talk, the current state of pre-surgical functional brain mapping and its evolution into an adaptive neurotechnology approach werediscussed. Examples of adaptive neurotechnology research conducted at the Functional Brain Mapping and Brain-Computer Interface Lab (FBM-BCI Lab), AdventHealth Orlando were presented, such as real-time functional brain mapping to prevent post-surgical deficits in epilepsy patients. Current gaps in this approach were demonstrated. The vision of integrating machine learning algorithms to advance functional brain mapping were provided. A potential application of machine learning for automatic brain signal classification will be discussed. The importance of multi-disciplinary efforts between neuroscientists and computer science, mathematics, and signal processing specialists were emphasized.


Here you will find information about the research publications authored by Dr. Milena Korostenskaja (Dr. K) - the Founder and Director of Research at The Institute of Neuroapproaches. Dr. K has established long-lasting fruitful collaborations with top scientists around the world. We are proud to showcase the results of this productive work with the list of publications below.




July 6, 2019

Clark, E., Czaplewski, A., Dourney, S., Gadelha, A., Nguyen, K., Pasciucco, P., Rios, M., Stuart, R., Castillo, E., Korostenskaja, M.

In: Communications in Computer and Information Science. 2019, CCIS Volume 1032, p. 243-254.

This work was prepared in collaboration with Dr. Elizabeth Clark and graduate students from the Department of Physical Therapy at Advent Health University. The project aims at improving functional motor outcomes in patients status post-stroke.


Stroke is the fifth leading cause of death and disability in the United States with approximately 6.8 million people living with residual deficits and approximately $34 billion spent on treatment annually. Simultaneously, dramatic healthcare shifts have limited extended care accessibility, with many individuals discharged from restorative therapy by three-months post-stroke. Decreased access and increased costs have led clinicians, and scientists to investigate more effective and efficient interventions to improve the function of the hemiparetic upper extremity of individuals post-stroke. One such modality is a brain-computer interface (BCI) technology that utilizes brain signals to drive rehabilitation of motor function. Emerging data suggests the use of BCI for motor rehabilitation post-stroke, facilitating an individual’s return to function and improving quality of life. Specifically, integration of virtual reality (VR) and functional electrical stimulation (FES) components is an innovative rehabilitation strategy with a strong potential to reinstitute central motor programs specific to hand function in patients’ status post-stroke. By utilizing the Fugl-Meyer Assessment (FMA), researchers can monitor the motor function of the hemiparetic upper extremity pre/post-intervention, objectively quantifying the effectiveness of BCI for the restoration of upper extremity motor function. Neurophysiological brain imaging techniques allow tracking changes in the neural substrates of motor function due to BCI intervention. Therefore, the purpose of our study is to demonstrate the utility of BCI-VR-FES intervention for motor rehabilitation of upper extremity, based upon the theory of neuroplasticity, in individuals’ post-stroke by using functional (FMA) and neurophysiological outcome measures.

July 9, 2019

Korostenskaja, M., RaviPrakash, H., Bagci, U., Lee, K.H., Chen, P.C., Salinas, C., Baumgartner, J., Castillo, E.M.

In: Brain-Computer Interface Research: A State-of-the-Art Summary 6, C. Guger, B. Allison, and J. Ushiba, Editors. 2019, Springer International Publishing, p. 11-29.

This work is a part of unique colalboratibe endavour with Artificial Intelligence in Medicine (AIM) specialists from the Center of Research in Computer Vision at the University of Central Florida - Dr. Ulas Bagci and his teammates. The project uses cutting-edge computer science approaches (machine learning, deep learning) to prevent functional morbidity of epilepsy surgery.


RATIONALE: Electrocorticography-based functional language mapping (ECoG-FLM) utilizes an ECoG signal paired with simultaneous language task presentation to create functional maps of the eloquent language cortex in patients selected for resective epilepsy or tumor surgery. At present, the concordance of functional maps derived by ECoG-FLM and electrical cortical stimulation mapping (ESM) remains rather low. This impedes the transition of ECoG-FLM into an independent functional mapping modality. As ESM is considered the gold standard of functional mapping, we aimed to use it in combination with machine learning (ML) approaches ("ESM-ML guide"), to improve the accuracy of ECoG-FLM.

METHODS: The ECoG data were collected from 6 patients (29.67±12.5 yrs; 19-52 yrs; 3 males, 3 females). Patient ECoG activity was recorded (g.USBamp, g.tec, Austria) during administration of language tasks. For data analysis: (1) All ECoG sites were divided into ESM positive (ESM (+)) and ESM negative (ESM(-)); (2) Features of ESM(+) and ESM(-) sites in the ECoG signal were determined by analyzing the signal in the frequency domain; (3) ML classifiers (Random Forest (RF) and Deep Learning (DL)) were trained to identify these features in language-related ECoG activity; (4) The accuracy of the ESM-ML guided classification was compared with the accuracy of the conventional ECoG-FLM.

RESULTS: The conventional approach demonstrated: 58% accuracy, 22% sensitivity, and 78% specificity.  The "ESM-ML guide" approach with RF classifier demonstrated: 76.2% accuracy, 73.6% sensitivity, and 78.78% specificity. The DL classifier achieved the highest performances compared to all others with 83% accuracy, 84% sensitivity and 83% specificity.

CONCLUSION: ECoG-FLM accuracy can be improved by using an "ESM-ML guide", making the use of ECoG-FLM feasible as a stand-alone methodology. The long-term goal is to create a tool-box with "ready to use an ESM-ML guide" algorithm trained to provide high accuracy ECoG-FLM results by classifying between ESM(+) and ESM(-) contacts in prospective sets of language-related ECoG data and, thus, contribute towards improved surgical outcomes.

February 8, 2019

Salillas, E., Korostenskaja, M., Kleineschay, T., Mehta, S., Vega, A., Castillo, E.

Frontiers in Psychology, section Cognitive Science, 2019, 10: 139 E-collection:

A common magnitude system for the processing of time and numerosity, supported by areas in the posterior parietal cortex, has been proposed by some authors. The present study aims to investigate possible intersections between the neural processing of non-numerical (time) and numerical magnitudes in the posterior parietal lobe. Using Magnetoencephalography for the comparison of brain source activations during the processing of duration and numerosity contrasts, we demonstrate parietal overlap as well as dissociations between these two dimensions. Within the parietal cortex, the main areas of overlap were bilateral precuneus, bilateral intraparietal sulci, and right supramarginal gyrus. Interestingly, however, these regions did not equivalently correlate with the behavior for the two dimensions: left and right precuneus together with the right supramarginal gyrus accounted functionally for durational judgments, whereas numerosity judgments were accounted by the activation pattern in the right intraparietal sulcus. Present results, indeed, demonstrate an overlap between the neural substrates for processing duration and quantity. However, the functional relevance of parietal overlapping areas for each dimension is not the same. In fact, our data indicate that the same parietal sites rule differently non-numerical and numerical dimensions, as parts of broader networks.


Making Headlines

January 3, 2020

Students, neurology residents, and doctoral students at the Faculty of Medicine of Vilnius University (VU) had the opportunity to discuss the latest topic in neurobiology, directly contributing to the successful execution of the Systems Biology program with a guest lecturer Dr Milena Korostenskaja, who is a much accomplished scientist in this field.
Professor Audronė Jakaitienė, coordinator of the Systems Biology Programme, also participated in the discussion – “The brain–computer interface and modelling of this interface were the subject of many questions and computational suggestions not only from students, but also from the Department of Statistics. VU scientists had the opportunity to broaden and deepen their knowledge of recent advances in computer-to-brain interface modelling. We are grateful to the VU Strategic Foundation, which contributed to the funding of this initiative,” she said.
After her visit, Dr Korostenskaja discussed the possibility of teaching in the Systems Biology programme again. Having this in mind, let us find out more about Dr Milena Korostenskaja...

Winter Springs, FL, USA

©2017 by Institute of Science and Education Innovations. Proudly created with