Institute of Fundamental Technological Research

Yap M., Bill C., Byra M., Ting-yu L., Huahu Y., Galdran A., Yung-Han C., Raphael B., Sven K., Friedrich C., Yu-wen L., Ching-hui Y., Kang L., Qicheng L., Ballester M., Carneiro G., Yi-Jen J., Juinn-Dar H., Pappachan J., Reeves N., Vishnu C., Darren D., Diabetic foot ulcers segmentation challenge report: Benchmark and analysis, Medical Image Analysis, ISSN: 1361-8415, DOI: 10.1016/j.media.2024.103153, Vol.94, No.103153, pp.1-14, 2024

Abstract:
Monitoring the healing progress of diabetic foot ulcers is a challenging process. Accurate segmentation of foot ulcers can help podiatrists to quantitatively measure the size of wound regions to assist prediction of healing status. The main challenge in this field is the lack of publicly available manual delineation, which can be time consuming and laborious. Recently, methods based on deep learning have shown excellent results in automatic segmentation of medical images, however, they require large-scale datasets for training, and there is limited consensus on which methods perform the best. The 2022 Diabetic Foot Ulcers segmentation challenge was held in conjunction with the 2022 International Conference on Medical Image Computing and Computer Assisted Intervention, which sought to address these issues and stimulate progress in this research domain. A training set of 2000 images exhibiting diabetic foot ulcers was released with corresponding segmentation ground truth masks. Of the 72 (approved) requests from 47 countries, 26 teams used this data to develop fully automated systems to predict the true segmentation masks on a test set of 2000 images, with the corresponding ground truth segmentation masks kept private. Predictions from participating teams were scored and ranked according to their average Dice similarity coefficient of the ground truth masks and prediction masks. The winning team achieved a Dice of 0.7287 for diabetic foot ulcer segmentation. This challenge has now entered a live leaderboard stage where it serves as a challenging benchmark for diabetic foot ulcer segmentation.

Keywords:
Deep learning, Diabetic foot ulcers, Segmentation, Convolutional neural networks

Gambin B. J., Kruglenko E., Tymkiewicz R., Litniewski J., Heating efficiency of agarose samples doped with magnetic nanoparticles subjected to ultrasonic and magnetic field, INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER, ISSN: 0017-9310, DOI: 10.1016/j.ijheatmasstransfer.2024.125467, Vol.226, No.125467, pp.1-10, 2024

Abstract:
Recently, magneto-ultrasound heating of tissue in the presence of magnetic nanoparticles (NPs) has been studied due to its high potential for use in oncological hyperthermia. It has been published that a synergistic effect, generation of additional heat caused by magneto-ultrasonic coupling, was observed in a tissue-mimicking material (TMM) enriched with magnetic NPs. The specific absorption rate (SAR) was determined from the temperature rise measurements in a focus of the ultrasound beam. It is important to use precise measurement methods when considering medical applications, for which there are limitations to the power of each field, resulting from the prevention of biological phenomena dangerous to the patient. This study demonstrates that in magneto-ultrasonic heating SAR can be measured much more accurately if the ultrasonic field is almost uniform. Measurements were performed on TMM containing Fe3O4 NPs with a diameter of approximately 8 nm and superparamagnetic properties. Both, the measurement and simulation results showed that the errors resulting from the inaccuracy of placing the temperature probe are smaller than in the case of the focused ultrasound. At the same time, the temperature increase caused by the ultrasonic field is almost linear and the influence of heat convection on the SAR determination is negligible. The measurements showed that magneto-ultrasonic hyperthermia can provide the desired thermal effect at lower ultrasound powers and magnetic fields compared to ultrasonic or magnetic hyperthermia used alone. No synergy effect was recorded.

Keywords:
Magnetic nanoparticle-mediated hyperthermia, Dual-mode ultrasonic-magnetic hyperthermia, Specific absorption rate, Hyperthermia efficiency

Pawłowska A., Ćwierz-Pieńkowska A., Domalik A., Jaguś D., Kasprzak P., Matkowski R., Fura , Nowicki A., Żołek N.S., Curated benchmark dataset for ultrasound based breast lesion analysis, Scientific Data, ISSN: 2052-4463, DOI: 10.1038/s41597-024-02984-z, Vol.11, No.148, pp.1-13, 2024

Abstract:
A new detailed dataset of breast ultrasound scans (BrEaST) containing images of benign and malignant lesions as well as normal tissue examples, is presented. The dataset consists of 256 breast scans collected from 256 patients. Each scan was manually annotated and labeled by a radiologist experienced in breast ultrasound examination. In particular, each tumor was identified in the image using a freehand annotation and labeled according to BIRADS features and lexicon. The histopathological classification of the tumor was also provided for patients who underwent a biopsy.
The BrEaST dataset is the first breast ultrasound dataset containing patient-level labels, image-level annotations, and tumor-level labels with all cases confirmed by follow-up care or core needle biopsy result. To enable research into breast disease detection, tumor segmentation and classification, the BrEaST dataset is made publicly available with the CC-BY 4.0 license.