Innovative AI-Driven Method for Predicting Drug Reactivity Through Brain Wave Analysis and Deep Learning developed by Lisorius for TIPS

Lisorius Secures Funding to Advance Brain Wave-Based Drug Reactivity Prediction

November 6, 2025

Seoul, South Korea – Lisorius, a pioneering technology firm specializing in psychiatric and neurological solutions, announced today its selection for the prestigious TIPS Global Track program. this crucial funding will accelerate the company’s Research and Advancement initiatives and facilitate expansion into international markets.

The company is currently engaged in over ten clinical research and collaborative development projects with leading university hospitals across Korea. Lisorius is also actively forging partnerships with research institutions and healthcare organizations in the United States, Japan, and Vietnam, signaling a widespread global ambition.

Addressing a Critical Challenge in Healthcare

Lisorius is focused on resolving a long-standing problem within psychiatry and neurology: accurately predicting how patients will react to medication. Their innovative approach leverages advanced brain wave analysis-specifically electroencephalography (EEG)-to identify patterns indicative of drug responsiveness.

While EEG is a non-invasive method for measuring brain activity, its widespread adoption has been hindered by the complexity of data pre-processing and standardization.Lisorius states it has overcome these hurdles with proprietary technology,creating a pathway to commercial viability.

Deep Learning and Explainable AI at the Core

The core of Lisorius’ technology lies in the application of deep learning and explainable artificial intelligence (XAI) to EEG data. This process allows for the extraction of significant details and the identification of digital biomarkers from the intricate signals generated by the brain.

The resultant deep learning technology aims to create a clinical decision support system. This system will provide medical professionals with interpretable insights by recognizing predictive features within brain wave frequency patterns, and visually demonstrating the supporting evidence. Early testing shows superior performance compared to existing EEG analysis models.

Initial focus areas include epilepsy, depression, and sleep disorders. The company anticipates its work will ultimately contribute to the realization of precision psychiatry and neurology.

NVIDIA Inception Program Partnership

Further accelerating its progress, Lisorius has joined the NVIDIA Inception program. This partnership grants access to high-performance GPU infrastructure, empowering the company to refine its EEG AI models and build a robust global medical AI ecosystem.This collaboration enhances the speed and efficiency of analyzing large EEG datasets.

“Brain waves hold immense potential as biosignals reflecting brain state,” stated Woo-seok Jeong, Director of Research and Development at Lisorius. “Our technology’s ability to purify and interpret this data will increase the clinical reliability of EEG and allow us to become a global leader in brain-based precision medicine.”

Did You Know? According to a report by Grand View Research, the global digital health market size was valued at USD 175.0 billion in 2021 and is expected to grow at a compound annual growth rate (CAGR) of 28.5% from 2022 to 2030, highlighting the increasing demand for innovative healthcare solutions.

Pro Tip: Understanding your brain wave patterns can be the first step to personalized health solutions. Consult with a neurologist or psychiatrist to discuss the potential benefits of EEG-based diagnostics.

Feature	Lisorius Technology	Customary EEG Analysis
Data Pre-processing	Automated & Standardized	Manual & Variable
AI Integration	Deep Learning & XAI	Limited
Clinical Support	Decision Support System	Primarily Diagnostic

The Future of Personalized Medicine

The advancement of brain wave analysis and AI-driven diagnostics represents a significant step toward personalized medicine. Tailoring treatments to individual physiological characteristics-like brain activity patterns-can dramatically improve outcomes and minimize adverse effects. This approach is poised to revolutionize how neurological and psychiatric disorders are diagnosed and managed.

The growing field of neurotechnology is also gaining traction, with advancements in brain-computer interfaces and non-invasive brain stimulation techniques. These innovations, coupled with AI-driven analysis, may unlock unprecedented possibilities for treating a wide range of conditions.

Frequently Asked Questions

• What is EEG analysis? EEG analysis measures electrical activity in the brain using electrodes placed on the scalp. It’s a non-invasive method used to diagnose various neurological conditions.
• How does Lisorius’ technology improve EEG analysis? Lisorius’ technology automates data pre-processing and applies deep learning and explainable AI to extract meaningful insights from EEG data.
• What are the target diseases for lisorius? Initially, the company is focused on epilepsy, depression, and sleep disorders, but aims to expand into other neurological and psychiatric conditions.
• What is the NVIDIA Inception Program? It’s a program that supports startups developing cutting-edge AI and data science technologies by providing access to resources and expertise.
• How will this technology improve drug responsiveness prediction? By analyzing brain wave patterns, the technology identifies biomarkers that can predict how a patient will respond to specific medications.
• What is Explainable AI (XAI)? XAI refers to AI models that provide understandable explanations for their outputs, increasing trust and transparency in clinical decision-making.
• What is the potential impact of precision psychiatry? Precision psychiatry aims to tailor treatments to individual patients based on their unique biological and clinical characteristics, potentially leading to more effective and personalized care.

What are your thoughts on the potential of AI in revolutionizing mental healthcare? Share your comments below!

What specific deep learning architecture demonstrated teh highest predictive accuracy for drug reactivity in Lisorius’s research, and why?

Understanding the Core Innovation: Neuro-pharmacogenomics

Recent advancements spearheaded by Lisorius and published in Trends in Pharmacological Sciences (TIPS) detail a groundbreaking approach to predicting individual drug reactivity. This isn’t simply another pharmacogenomic study; it’s a paradigm shift leveraging the correlation between brain wave patterns – specifically electroencephalography (EEG) data – and a patient’s response to pharmaceutical interventions. This novel field, termed neuro-pharmacogenomics, utilizes artificial intelligence (AI) and deep learning algorithms to decode these complex neurological signals.

The Role of EEG in Drug Response prediction

Traditionally, predicting drug response relies heavily on genetic markers, metabolic enzyme activity, and clinical factors. However, these methods often fall short, explaining only a fraction of the observed variability in patient outcomes. Lisorius’s research highlights the brain’s inherent electrical activity as a crucial, previously untapped biomarker.

* EEG as a Dynamic Biomarker: EEG captures real-time brain activity, reflecting complex neuronal interactions influenced by genetics, habitat, and current physiological state.

* Individualized Brain Signatures: Each individual exhibits a unique EEG signature, even before drug administration.

* Drug-Induced EEG Changes: Pharmacological agents demonstrably alter these EEG patterns, providing a window into the drug’s mechanism of action within the patient’s brain.

Deep Learning Architectures for EEG Data analysis

The sheer volume and complexity of EEG data necessitate sophisticated analytical tools. lisorius’s team employs several deep learning architectures, including:

1. Convolutional Neural networks (CNNs): Effective at identifying patterns in time-series data like EEG signals, CNNs can automatically extract relevant features without manual engineering.
2. Recurrent Neural Networks (RNNs), especially LSTMs: RNNs excel at processing sequential data, capturing the temporal dependencies within EEG recordings. Long Short-Term Memory (LSTM) networks address the vanishing gradient problem, allowing them to learn long-range dependencies.
3. Hybrid Models: Combining CNNs and RNNs often yields superior performance, leveraging the strengths of both architectures. Such as, a CNN might extract spatial features from EEG channels, while an LSTM analyzes the temporal evolution of these features.

Data Preprocessing and Feature Engineering

Before feeding EEG data into the AI models, rigorous preprocessing is essential. This includes:

* Artifact Removal: Eliminating noise from eye blinks, muscle movements, and electrical interference. Techniques include Autonomous Component Analysis (ICA) and filtering.

* Data Segmentation: Dividing continuous EEG recordings into epochs relevant to drug administration and subsequent response assessment.

* Feature Extraction: Calculating relevant features from the EEG signal, such as power spectral density (PSD) in different frequency bands (delta, theta, alpha, beta, gamma), coherence, and entropy. Machine learning algorithms benefit from well-defined features.

Predicting Drug Reactivity: A Step-by-Step Process

Lisorius’s method involves a multi-stage process:

1. Baseline EEG Recording: Obtain a baseline EEG recording from the patient before drug administration.
2. Drug Administration & Monitoring: Administer the drug and continuously monitor EEG activity.
3. Data Acquisition & Preprocessing: acquire and preprocess the EEG data as described above.
4. Model Training & Validation: train the deep learning model on a large dataset of EEG recordings and corresponding drug response data (e.g., therapeutic efficacy, adverse effects). Rigorous validation using independent datasets is crucial to prevent overfitting.
5. Prediction & Personalization: Input the patient’s EEG data into the trained model to predict their likely response to the drug. This allows for personalized medicine approaches, tailoring treatment regimens to individual neurological profiles.

Applications Beyond Traditional Pharmacology

The implications of this research extend far beyond simply improving drug selection. Potential applications include:

* Neurological Disorder Diagnosis: Identifying subtle EEG abnormalities associated with specific neurological conditions.

* anesthesia Depth Monitoring: Optimizing anesthesia dosage based on real-time EEG feedback.

* Predicting Response to Neuromodulation Therapies: Determining which patients are most likely to benefit from treatments like Transcranial Magnetic Stimulation (TMS) or Deep Brain Stimulation (DBS).

* Clinical Trial Optimization: Stratifying patients in clinical trials based on their predicted drug response, increasing the likelihood of detecting true treatment effects. Pharmacogenomics plays a key role here.

Benefits of AI-Driven Drug Reactivity Prediction

* Reduced Adverse Drug Reactions (ADRs): By identifying patients at high risk of adverse effects, clinicians can avoid prescribing potentially harmful medications.

* Improved Treatment Efficacy: Selecting the most effective drug for each patient maximizes the chances of a positive therapeutic outcome.

*