We are pleased to announce that our research article on the development of a method for calculating ultrasound treatment parameters using neural networks has been published in an international journal.
This method opens new horizons in the field of medicine, making treatment procedures for tremor and Parkinson’s disease using MRI-guideded focused ultrasound (MRgFUS) more predictable, safe, and effective.
Why is it so important to accurately control these parameters?
We pay special attention to precise planning, and three key parameters of ultrasound impact play a special role in achieving the optimal result:
Insufficient power, too short duration, or inadequate energy may result in the tremor not being reduced sufficiently. On the other hand, excessive power, too long exposure, or too high energy may damage surrounding tissues and cause undesirable side effects.
During operations using ultrasound, precise temperature regulation is a key factor for achieving a successful outcome. However, numerous variables, including bone density and individual patient characteristics, make this process complex. We have developed a neural network model that allows for highly accurate prediction of exposure parameters, such as power, duration, and energy, to achieve the necessary temperature in the target area.
Our approach is based on the analysis of data from more than 150 procedures. Using advanced machine learning technologies, we have created a model that predicts temperature with a minimal error of only 1.93° C. It is more accurate than traditional linear methods and provides a high level of reliability and predictability in treatment.
“Our goal is to make the treatment of tremor and Parkinson’s disease using MRI-guided focused ultrasound more predictable, safe, and effective for all our patients ” – neurologist at V.S. Buzaev International Medical Centre (Ufa), external PhD student at the Scientific Center of Neurology (Moscow) Alsu Narkisovna Khatmullina
Our method allows doctors to calculate the optimal parameters of ultrasound exposure in advance, taking into account key factors such as age, gender, bone density, and tissue response to the first exposure. This means that it is now possible to minimize risks, increase the effectiveness of procedures, and provide patients with safer treatment.
We see the future of medicine in a personalized approach, where artificial intelligence technologies serve as a reliable tool in the hands of the physician. Our developed model is an example of how, based on accurate data and machine learning, the effectiveness and safety of tremor treatment can be significantly enhanced. We believe that our research will form the foundation for further development of ultrasound treatment methods and help thousands of patients worldwide.
We are proud to have contributed to the development of medicine and thank everyone who supported us on this journey. Ahead of us are new challenges and discoveries that will make medical technologies even more accessible and effective.
Join us on the path to innovation in medicine!
Full text of the article can be read here:
Alsu Narkisovna Khatmullina, Diana Shamilevna Avzaletdinova, Dinara Ilgizovna Nabiullina, Sergey Nikolaevich Illarioshkin, Guzaliya Minvazykhovna Sakharova, Naufal Shamilevich Zagidullin, Nadezhdina Ekaterina Andreevna, Shamil Makhmutovich Safin, Rezida Maratovna Galimova. Predicting magnetic resonance imaging-guided focused ultrasound sonication parameters beyond skull density ratio. Global Translational Medicine 5419. https://doi.org/10.36922/gtm.5419
Doctor - Obstetrician-Gynecologist, Ultrasound Diagnostics Doctor