Advanced depression identifier tool

Innowise has developed an innovative AI platform that can identify depression in patients through EEG scans.

Customer

Industry

Medicine

Region

USA

Client since

2022

Our client is one of the major representatives in healthcare. They run their own medical center in the USA.

Detailed information about the client cannot be disclosed under the provisions of the NDA.

Challenge

Over 1 billion people worldwide suffer from mental disorders, with depression affecting more than 300 million. To aid in early diagnosis and comprehensive treatment, researchers have identified EEG biomarkers and AI facial emotion recognition technology as promising tools. By utilizing AI facial emotion recognition, which uses machine learning to analyze facial expressions and detect patterns associated with mental disorders, we can provide a non-invasive and convenient method for detecting potential mental health issues. With facial emotion recognition using machine learning, we can augment traditional clinical approaches to mental health diagnosis and treatment, providing more effective and inclusive solutions.

Innowise was approached by a client with the requirement of developing an automated solution that utilizes AI to detect human emotions related to depression in patients. By leveraging advanced emotion AI technologies and expertise, Innowise developed a solution that can assist clinicians in providing timely and effective care to those struggling with depression.

Solution

We implemented AI as a service solution to aid in the detection and treatment of depression. Using state-of-the-art deep learning tools, we developed a model that can detect depression by scanning EEG results and identify EEG predictors of therapeutic response. By analyzing EEG data, our model can identify patterns and indicators that may help clinicians tailor their treatment approach for each individual patient.

CLOUD ML APPLICATION

We opted for a cloud-based application since it provides a host of benefits for our machine learning (ML) solution, including enhanced security and data storage capabilities. The implemented SaaS solution eliminates the need for high processing power, data storage, and multiple servers to process ML algorithms simultaneously.

Our team has also developed an API that improves the user experience by automatically launching trained machine learning models to process user data and show real-time results.

Overall, the developed cloud-based SaaS solution and accompanying API provide a comprehensive and streamlined approach to machine learning, empowering our clients with the capabilities they need to achieve their objectives.

ML TRAINING

To support our AI models and predictive analytics, our development team implemented data lakes, providing a robust and scalable storage solution for big volumes of data. This enables us to conduct extensive emotional AI analysis and extract valuable insights for our clients. Then we seamlessly integrated data warehouses, complete the transformation process and effectively cleanse the data before uploading it.

When the EEG scan reaches the cloud, the ML model harnesses the data stored in the data warehouse to assess and accurately determine if the patient has depression.

It is important to note that working with medical data was the most challenging part of the development. However, the Innowise team managed to successfully train the ML model and integrate it into medical practice.

This achievement not only demonstrates our team’s proficiency in handling complex and sensitive medical data but also highlights our commitment to providing the best possible solutions for our clients.

WEB INTERFACE

To simplify the process of getting results, we developed an intuitive web interface that streamlines the user experience. This solution eliminates the need for manual data entry, significantly reducing the risk of errors and allowing users to easily and quickly obtain accurate and reliable results.

Moreover, thanks to the intuitive interface, it is to navigate through the system and obtain the necessary data without any technical expertise or complex procedures.

Technologies

Back-end

Python, FastAPI

Machine learning

OpenCV, Pandas, NumPy, Scikit-learn, PyTorch, Matplotlib, MLFlow, Keras, Tensorflow, Python-MIP

Cloud & DevOps

AWS(S3, Lambda, SageMaker etc.), Kubernetes, Docker

Process

Despite the complex and multi-step development process, the Innowise team had enough expertise to address all issues and problems timely.

In the first stage, we engaged a specialist in model validation who used various tools for ML exploring ML model predictions. Huge efforts were given for a thorough data labeling preparation, which ultimately led to enormous time savings since we had configured convenient infrastructure for all specialists. The research step included various model probations and was efficiently conducted via a designed validation schema.

After our specialists filtered the data, they began training the ML model. This phase consisted of several stages of improving and refining the model. Finally, developers integrated the trained model into the cloud application.

As for project management, we used Slack and Jira to collaborate on the project within the company and Google Chats for external communication with the client. We utilized Scrum methodology, with daily meetings and demo presentations of intermediate results every month.

As of today, we continue to support the project and resolve any issues that arise until everything works properly on the client’s side.

Team

Back-End Developer

Front-End Developer

Data Scientist

Machine Learning Engineer

Business Analysts

Designer

Project Manager

QA Engineer

Results

Our team delivered an advanced AI mental health app to our client, providing them with a trained model capable of detecting depression from EEG scans and identifying biomarkers for predicting treatment response. This innovative ML platform is a novel approach to treating depression that increases the likelihood of new drug approval.

The designed AI-based mental health app is easy to use for medical professionals since the scanned results are managed through an intuitive web interface. Moreover, the development team built a data collection system with a toolkit for quick data labeling, optimizing the process for clinicians and researchers.

Since implementing the designed solution, the customer has seen significant benefits, including increased clinic funds and an expanded client base. By offering a unique tool for depression treatment, our client has positioned themselves at the forefront of the industry and attracted more patients seeking cutting-edge treatments.

Project duration

3 months for MVP
The project is still ongoing; at this stage we provide maintenance and support to the developed platform

Related cases

Contact us!

Book a call or fill out the form below and we’ll get back to you once we’ve processed your request.

What happens next?

Having received and processed your request, we will get back to you shortly to detail your project needs and sign an NDA to ensure the confidentiality of information.

After examining requirements, our analysts and developers devise a project proposal with the scope of works, team size, time, and cost estimates.

We arrange a meeting with you to discuss the offer and come to an agreement.

We sign a contract and start working on your project as quickly as possible.

Advanced depression identifier tool

Customer

Challenge

Solution

Technologies

Process

Team

Results

Related cases

Contact us!

What happens next?

Let^●s develop
software together!

Close

Advanced depression identifier tool

Customer

Challenge

Solution

Technologies

Process

Team

Results

Related cases

Contact us!

What happens next?

Let●s develop software together!

Subscribe to our newsletter

Close

Let^●s develop
software together!

Subscribe
to our newsletter