Pavel-Dumitru Cernelev

@article{hashtrudi-zaad2024,

title = {Expert-guided optimization of ultrasound segmentation models for 3D spine imaging},

author = {Kian Hashtrudi-Zaad and Tamas Ungi and Chris Yeung and Zachary Baum and Pavel-Dumitru Cernelev and Anthony N Hage and Christopher Schlenger and Gabor Fichtinger},

year  = {2024},

date = {2024-01-01},

pages = {680-685},

publisher = {IEEE},

abstract = {We explored ultrasound for imaging bones, specifically the spine, as a safer and more accessible alternative to conventional X-ray. We aimed to improve how well deep learning segmentation models filter bone signals from ultrasound frames with the goal of using these segmented images for reconstructing the 3-dimensional spine volume.Our dataset consisted of spatially tracked ultrasound scans from 25 patients. Image frames from these scans were also manually annotated to provide training data for image segmentation deep learning. To find the optimal automatic segmentation method, we assessed five different artificial neural network models and their variations by hyperparameter tuning. Our main contribution is a new approach for model selection, employing an Elo rating system to efficiently rank trained models based on their visual performance as assessed by clinical users. This method addresses the …},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{cernelev2024,

title = {Evaluation of an Implantable Electromagnetic Microsensor for Computer-Assisted Surgery},

author = {Pavel-Dumitru Cernelev and Leah Groves and Gernot Kronreif and Tamás Ungi and Gábor Fichtinger},

year  = {2024},

date = {2024-01-01},

pages = {212-216},

abstract = {Computer-assisted surgical navigation systems require a high tracking accuracy while not occupying much space. Currently, the size of the electromagnetic tracking sensors in use can be distracting to the surgeon. An electromagnetic microsensors (1.04 mm x 7.9 mm) has been developed to promote seamless integration within surgical workspace. This study evaluates the performance accuracy of this microsensors, to determine the suitability for the clinical setting. A series of experiments were performed to determine the sensor’s accuracy in a controlled environment, in the presence of ferromagnetic materials, and while an electrocautery is used. The electromagnetic sensor was compared to an optical tracking ground truth to determine the tracking error. Initial tests in the simulated surgical environment demonstrate that the microsensor maintains a below 1 mm error and minimal jitter error when 15 cm away from the field generator. However, accuracy decreases in the presence of ferromagnetic materials and an electrocautery, especially if the electrocautery is in coagulation mode which can result in sensor damage. These findings highlight the sensor’s potential in surgical navigation, while also indicating the need for further improvements to ensure functionality in the presence of surgical tools and in varying operational conditions.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}