Brain imaging data collected for a quarter of a century is in artificial intelligence!

Haber ile ilişkili SDG etiketleri

DOI : https://doi.org/10.32739/uha.id.44352

Neuroimaging (EEG and fMRI) data obtained at NPİstanbul Hospital for 26 years were analyzed in the application and research centers of Üsküdar University and the BraiNP/NP Model was created. The model, which uses Artificial Intelligence (AI) algorithms, provides a preliminary diagnosis of different psychiatric disorders.

Stating that BraiNP was developed under the consultancy of Prof. Nevzat Tarhan and made available via web interface at npmodel.com, Head of Software Engineering Department Prof. Türker Tekin Ergüzel said that "In its current form, BraiNP provides high accuracy with predictive models of obsessive compulsive disorder (OCD), healthy control, unipolar-bipolar and transcranial magnetic stimulation (TMS) response in depression."

Üsküdar University Advisor to the Rector, Faculty of Engineering and Natural Sciences (FENS) Head of Software Engineering Department Prof. Türker Tekin Ergüzel informed on the BraiNP/NP Model developed under the consultancy of Prof. Nevzat Tarhan.

Neuroimaging data collected since 1998 were classified with artificial intelligence

Informing on the system called BrainNP or NP Mode Prof. Türker Tekin Ergüzel made the following statement:

"NP Model is a model with high predictive skills developed by analyzing the neuroimaging (EEG and fMRI) data collected at NPİstanbul Hospital with its international experience in the diagnosis and treatment of psychiatric diseases since its establishment in 1998 and using Artificial Intelligence (AI) algorithms in all processes and developed for the pre-diagnosis classification or treatment outcome prediction of different psychiatric diseases."

Goal is to bring the collected data to the health system

Prof. Ergüzel explained the goal of the model as follows: "The prediction models of this model, which were previously carried out within NPİstanbul and Üsküdar University, are not limited to scientific publications, and the collected data is brought back to the health system and aims to use the resources of physicians, clients and health systems effectively in the early pre-diagnosis and treatment outcome prediction processes of diseases."

"The basis of the developments is also the increasing resolution of the collected data"

Stating that in the last three years, there has been a significant development in classical artificial intelligence (AI) algorithms to classify diseases using biomarkers, Ergüzel noted that the basis of these developments is the increasing resolution of the collected data, the diversification of patient data sets, and especially the widespread use of deep learning algorithms.

Explaining that the new generation learning algorithms can successfully extract the distinctive features found in the raw data in the classification processes, especially data such as EEG with high temporal resolution and fMRI with high spatial resolution, after they are obtained from patients or healthy control groups, they are purified from noise with pre-processing steps. He noted that feature extraction is performed by computers with GPUs.

International patent application has been filed

Stating that the NP Model was developed under the consultancy of Prof. Nevzat Tarhan within the framework of a project supported within the scope of Scientific Research Projects of Üsküdar University and made available via web interface at npmodel.com, Prof. Türker Tekin Ergüzel continued his remarks as follows:

"In its current form, BraiNP provides high accuracy with predictive models of obsessive compulsive disorder (OCD), healthy control, unipolar-bipolar and transcranial magnetic stimulation (TMS) response in depression. Moreover, the system is designed with new data to make more stable predictions. The model, which was developed with the capacity of pre-diagnosis in the classification of common psychiatric disorders such as depression, OCD, ADHD, bipolar disorder, trichotillomania, and addiction, was designed together with neurologists and psychiatrists at NPİstanbul Hospital, neuroscience experts and software engineers at Üsküdar University. The model has been applied for an international patent. The patent registration is the registration of the potential of the application and its original, innovative skill, and it has been opened to the use of NPİstanbul Hospital physicians."

7 basic contributions will be made for patients, physicians and the health system

Stating that in this way, 7 basic contributions will be made for the patient, physician and health system in the short and long term, Prof. Türker Tekin Ergüzel listed them as follows:

"Early Intervention: Early detection of mental health issues allows for rapid intervention and treatment that can prevent the condition from getting worse. Early intervention is often associated with better treatment outcomes and a better prognosis.

Prevention of Complications: Detecting mental health disorders at an early stage helps prevent the development of complications such as comorbid conditions, substance abuse, or self-injurious behaviors.

Reduced Suffering: Timely diagnosis ensures that individuals receive appropriate support and treatment, reducing their suffering and improving their quality of life. It can alleviate symptoms and help individuals cope better with their condition.

Personalized Treatment Plans: Pre-diagnosis provides a basis for developing personalized treatment plans that are tailored to the individual's specific needs and circumstances. This approach increases the likelihood of treatment effectiveness and patient satisfaction.

Resource Allocation: Early detection allows for better allocation of resources within the healthcare system. It ensures that patients receive the appropriate level of care, reducing the burden on emergency departments and preventing unnecessary hospitalizations.

Education and Support: Early knowledge of the diagnosis allows individuals and their families to access relevant education and support services. This allows them to better understand the situation, learn coping strategies, and access community resources for ongoing support.

Improved Prognosis: With early detection and intervention, there is a better chance of managing symptoms effectively and improving the long-term prognosis. It can also minimize the risk of disease recurrence and facilitate recovery."

"Brain-computer interfaces may be useful for rehabilitation after stroke"

Stating that in addition to BCI (Brain-Computer Interfaces) and artificial intelligence studies, students are provided with application and clinical opportunities on subjects such as brain stimulation, neuro-imaging laboratories and health physics in health informatics, Prof. Türker Tekin Ergüzel continued his remarks as follows:

"Brain-computer interfaces receive brain signals, analyze them, and convert them into commands that are sent to output devices that perform desired actions. The main function of the BCI is to replace or restore the beneficial functions of patients with disabilities due to neuromuscular disorders such as amyotrophic lateral sclerosis, cerebral palsy, stroke, or spinal cord injury.

Brain-computer interfaces may also be useful for post-stroke rehabilitation and other disorders. Our neuroscience research, which is at the center of developments, offers researchers the opportunity to develop applications with Neuroscience Master's and Doctorate programs in our graduate programs."

Üsküdar News Agency (ÜNA)