Article
Oncology
Diana Veiga-Canuto, Leonor Cerda-Alberich, Cinta Sanguesa Nebot, Blanca Martinez de Las Heras, Ulrike Potschger, Michela Gabelloni, Jose Miguel Carot Sierra, Sabine Taschner-Mandl, Vanessa Duster, Adela Canete, Ruth Ladenstein, Emanuele Neri, Luis Marti-Bonmati
Summary: Tumor segmentation is a crucial step in imaging processing and is usually done manually by radiologists. This study developed an automatic model using the deep learning architecture nnU-Net to detect and segment neuroblastic tumors, and compared its performance with manual segmentation. The results showed that the variability between nnU-Net and manual segmentation is similar to the inter-observer variability in manual segmentation. Additionally, the automatic model demonstrated a significant time advantage over manual segmentation.
Article
Oncology
H. Petra Kok, Johannes Crezee
Summary: This study developed a temperature-based re-optimization strategy and compared its predicted effectiveness with clinically applied protocol/experience-based steering. The results showed that the re-optimization strategy effectively suppressed hot spots without affecting tumor temperatures, similar to clinical steering. Implementing these strategies can significantly improve the overall treatment quality within hyperthermia centers.
Article
Veterinary Sciences
Asger L. Jensen, Anne K. H. Krogh, Lise N. Nielsen
Summary: This study assessed the correlation between the grading of anisocytosis in canine blood smears and RDW, as well as the intra- and inter-observer variation in assessing anisocytosis. The results showed a significant correlation between anisocytosis graded on blood smears and RDW values. The weak to moderate observer agreement in the visual assessment of anisocytosis suggests that RDW measurement should be used for clinical decision-making.
FRONTIERS IN VETERINARY SCIENCE
(2023)
Article
Physics, Condensed Matter
M. Roustaei, M. Servatkhah
Summary: This study investigates the temperature distribution within a tumor after the injection of magnetic nanoparticles using a theoretical model and numerical simulations. The results show that the temperature distribution depends on particle size, injection concentration, and dose, and the optimal hyperthermia results can be achieved by considering intermediate radius nanoparticles, high injection dose, and concentrations.
PHYSICA B-CONDENSED MATTER
(2022)
Article
Radiology, Nuclear Medicine & Medical Imaging
Jingwei Duan, Mark Bernard, Laura Downes, Brooke Willows, Xue Feng, Waleed F. Mourad, William St Clair, Quan Chen
Summary: This study evaluates the performance of a commercial AI-based contouring model in the automated prostate treatment planning process. The results demonstrate good geometric accuracy and clinical acceptability of the AI contours, and the automated treatment plans based on these contours are comparable to those based on reference contours.
Article
Oncology
Adela Ademaj, Emsad Puric, Dietmar Marder, Olaf Timm, Thomas Kern, Roger A. Halg, Susanne Rogers, Oliver Riesterer
Summary: Radiotherapy combined with deep regional hyperthermia is a promising organ-preservation therapy for elderly and frail patients with muscle-invasive bladder cancer, as it can effectively control local progression of the tumor while avoiding systemic toxicities.
INTERNATIONAL JOURNAL OF HYPERTHERMIA
(2023)
Article
Engineering, Biomedical
Chenggang Lu, Jinli Yuan, Kewen Xia, Zhitao Guo, Muxuan Chen, Hengyong Yu
Summary: In this study, a regional perception and multi-scale feature fusion network (RMFNet) is proposed for cardiac magnetic resonance (CMR) image segmentation, which can accurately segment cardiac images. The results show that RMFNet outperforms other advanced methods in performance, reducing the stress of radiologists and playing an important role in image-guided clinical procedures.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Anjali Balagopal, Howard Morgan, Michael Dohopolski, Ramsey Timmerman, Jie Shan, Daniel F. Heitjan, Wei Liu, Dan Nguyen, Raquibul Hannan, Aurelie Garant, Neil Desai, Steve Jiang
Summary: The study investigated automatic segmentation of medical images using deep learning algorithms, aiming to determine the consistency and impact of physician styles on treatment outcomes. Results showed that physician styles are learnable and do not significantly affect treatment outcomes. A physician style-aware segmentation network was developed to improve segmentation accuracy.
ARTIFICIAL INTELLIGENCE IN MEDICINE
(2021)
Article
Computer Science, Artificial Intelligence
Lidong Yang, Jialin Jiang, Xiaojie Gao, Qinglong Wang, Qi Dou, Li Zhang
Summary: This study presents a real-time autonomous distribution planning method based on deep learning for controlling the shape, position, and navigation of microrobot swarms in complex environments. It proposes a framework with different autonomy levels for environment-adaptive navigation and demonstrates the capability of autonomous swarm navigation for targeted delivery and cargo transport in environments with channels or obstacles.
NATURE MACHINE INTELLIGENCE
(2022)
Article
Engineering, Biomedical
Xuan Chen, Boyang Zhou, Linfei Xiong, Chongke Zhao, Lifan Wang, Yongwei Zhang, Huixiong Xu
Summary: This study introduces an interactive segmentation framework with only four clicks for precise segmentation of breast lesions, utilizing RoI focusing module and RoI & Global re-calibration module. Experimental results demonstrate high accuracy achieved by the proposed framework on an unseen test set, showing the effectiveness of the method.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Biochemical Research Methods
David Borland, Carolyn M. McCormick, Niyanta K. Patel, Oleh Krupa, Jessica T. Mory, Alvaro A. Beltran, Tala M. Farah, Carla F. Escobar-Tomlienovich, Sydney S. Olson, Minjeong Kim, Guorong Wu, Jason L. Stein
Summary: Segmentor is an open-source tool designed for reliable, efficient, and user-friendly manual annotation and refinement of objects within 3D light sheet microscopy images, particularly nuclei. It employs a hybrid 2D-3D approach for visualizing and segmenting objects, with features for automatic region splitting to streamline the process of 3D segmentation. Editing simultaneously in 2D and 3D using Segmentor significantly decreases time spent on manual annotations without affecting accuracy.
BMC BIOINFORMATICS
(2021)
Article
Oncology
Alvaro Yago Ruiz, Marta Cavagnaro, Lorenzo Crocco
Summary: Non-invasive temperature monitoring is crucial for hyperthermia cancer treatment to ensure the therapeutic temperature is reached. Current methods used are either superficial or invasive. This study proposes the use of deep learning with microwave imaging for temperature monitoring, showing promising results with an accuracy above 90% in the case of neck tumors.
Article
Agriculture, Multidisciplinary
Qilin Sun, Ruirui Zhang, Liping Chen, Linhuan Zhang, Hongming Zhang, Chunjiang Zhao
Summary: A new semantic segmentation network called SADNet was proposed for segmenting orchard UAV images. It achieved better performance compared to other models on the orchard dataset and performed well on the public dataset as well. Additionally, a method for constructing orchard raster maps based on the segmentation results was proposed, which can be used for planning intra- and inter-row paths in orchards.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Oncology
Takayuki Ohguri, Kagayaki Kuroda, Katsuya Yahara, Sota Nakahara, Sho Kakinouchi, Hirohide Itamura, Takahiro Morisaki, Yukunori Korogi
Summary: Several studies have shown that concurrent use of hyperthermia with radiotherapy can significantly increase local control rates, but achieving a favorable increase in tumor temperature is more challenging in deep-seated tumors. This study investigated methods to reduce subcutaneous fat overheating and optimize temperature distribution in deep regional hyperthermia in the pelvis using electromagnetic field numerical simulations. The use of overlay boluses and salt solution concentration were found to be important for optimizing temperature distribution.
Article
Chemistry, Analytical
Gulsah Yildiz, Halimcan Yasar, Ibrahim Enes Uslu, Yusuf Demirel, Mehmet Nuri Akinci, Tuba Yilmaz, Ibrahim Akduman
Summary: This paper proposes an antenna excitation optimization scheme applicable to various configurations and investigates it using CNN-based approaches. The results demonstrate that this method outperforms traditional lookup table methods. The proposed deep-learning-based optimization technique holds great application potential in microwave hyperthermia.
Article
Oncology
Tim-Oliver Sauer, Oliver J. Ott, Godehard Lahmer, Rainer Fietkau, Christoph Bert
Summary: A markerless workflow for breast cancer patients has been introduced and evaluated retrospectively. The results show that this workflow can save time and dose while ensuring appropriate positioning accuracy.
STRAHLENTHERAPIE UND ONKOLOGIE
(2023)
Article
Oncology
Adela Ademaj, Paraskevi D. Veltsista, Dietmar Marder, Roger A. Haelg, Emsad Puric, Thomas B. Brunner, Hans Crezee, Dorota Gabrys, Martine Franckena, Cihan Gani, Michael R. Horsman, Robert Krempien, Lars H. Lindner, Sergio Maluta, Markus Notter, Griseldis Petzold, Sultan Abdel-Rahman, Antonella Richetti, Andreas R. Thomsen, Pelagia Tsoutsou, Rainer Fietkau, Oliver J. Ott, Pirus Ghadjar, Oliver Riesterer
Summary: There is significant heterogeneity among European clinical centers in the practice of hyperthermia, including the indications treated and the recording of thermometric parameters. More evidence from clinical studies is needed to standardize hyperthermia practice.
STRAHLENTHERAPIE UND ONKOLOGIE
(2023)
Article
Oncology
Oliver J. Ott, Udo S. Gaipl, Allison Lamrani, Rainer Fietkau
Summary: For decades, the antineoplastic potential of hyperthermia alone or in combination with radiotherapy and/or chemotherapy has been extensively researched in various tumor entities. This article summarizes the clinical evidence on the beneficial effects of additional hyperthermia in combination with intravesical Mitomycin C for superficial non-muscle-invasive bladder cancer, as well as deep regional microwave hyperthermia techniques used during external beam radiotherapy or chemoradiation for more advanced tumors. The presented data justifies further prospective trials on multimodality treatment for bladder preservation, including mild regional hyperthermia.
SEMINARS IN RADIATION ONCOLOGY
(2023)
Review
Oncology
Thomas Weissmann, Michael Rueckert, Florian Putz, Anna-Jasmina Donaubauer, Markus Hecht, Soren Schnellhardt, Philipp Schubert, Johannes Roesch, Daniel Hoefler, Oliver J. Ott, Marlen Haderlein, Sebastian Lettmaier, Rainer Fietkau, Benjamin Frey, Udo S. Gaipl, Lisa Deloch
Summary: Osteoarthritis is a common and economically relevant disease, especially in an ageing population. Despite numerous conventional treatment options, many patients are still refractory to treatment. Low-dose radiotherapy has shown potential in treating patients with inflammatory and degenerative diseases, but its effectiveness has not been proven in randomized trials. Further research is needed to investigate the underlying mechanisms and optimize patient stratification and treatment procedures.
STRAHLENTHERAPIE UND ONKOLOGIE
(2023)
Article
Cell Biology
Mareike F. Bauer, Laura S. Hildebrand, Marie-Charlotte Rosahl, Ramona Erber, Soren Schnellhardt, Maike Buettner-Herold, Florian Putz, Oliver J. Ott, Carolin C. Hack, Rainer Fietkau, Luitpold Distel
Summary: Cell-in-cell (CIC) structures in early hormone-sensitive breast cancer were studied. In vitro experiments showed that breast cancer cell lines had higher phagocytic capacity compared to primary human fibroblasts. Tissue analysis from 147 patients in a clinical trial revealed that CIC structures were associated with favorable local recurrence-free survival and disease-free survival, but unfavorable for metastasis-free survival. CIC had the highest prognostic value in young breast cancer patients.
Article
Oncology
Andre Karius, Juliane Szkitsak, Vratislav Strnad, Michael Lotter, Stephan Kreppner, Philipp Schubert, Rainer Fietkau, Christoph Bert
Summary: This study aimed to evaluate implant stability and identify causes of implant variations during high-dose-rate multi-catheter breast brachytherapy. The findings showed that multi-catheter breast brachytherapy has high implant stability overall, but changes in skin dose should be considered.
RADIOTHERAPY AND ONCOLOGY
(2023)
Article
Oncology
Andre Karius, Claudia Schweizer, Vratislav Strnad, Michael Lotter, Stephan Kreppner, Allison Lamrani, Rainer Fietkau, Christoph Bert
Summary: The purpose of this study was to investigate differences in seed-displacements between the immediate post-implant phase and the time to post-plan CT in seed prostate brachytherapy. Seed positions were analyzed on treatment plans and CT scans for 33 patients, and the resulting dosimetric effects were calculated. The study found that seed-displacements were greater in the immediate post-implant phase compared to the time to post-plan CT, with enhanced displacements in the superior-inferior direction.
RADIOTHERAPY AND ONCOLOGY
(2023)
Article
Oncology
Oliver J. Ott, Wilhelm Stillkrieg, Ulrike Lambrecht, Claudia Schweizer, Allison Lamrani, Tim-Oliver Sauer, Vratislav Strnad, Christoph Bert, Carolin C. Hack, Rainer W. Fietkau, Matthias W. Beckmann
Summary: Compared to whole-breast irradiation, partial-breast irradiation reduces radiation doses to healthy tissues and should be recommended to suitable patients to minimize risks of secondary tumors and major cardiac events.
Article
Oncology
Tim-Oliver Sauer, Wilhelm Stillkrieg, Oliver J. Ott, Rainer Fietkau, Christoph Bert
Summary: The purpose of this study was to determine the maximum allowed shift deviations in six degrees of freedom for surface guided radiation therapy of breast cancer patients. The robustness of treatment plans against 6DOF shifts was investigated, and threshold values were identified based on dose constraints. The study emphasized the need to consider differences between DOFs and asymmetrical sensitivities when evaluating the dosimetric impact of position deviations.
RADIATION ONCOLOGY
(2023)
Article
Oncology
Thomas Weissmann, Sina Mansoorian, Matthias Stefan May, Sebastian Lettmaier, Daniel Hoefler, Lisa Deloch, Stefan Speer, Matthias Balk, Benjamin Frey, Udo S. Gaipl, Christoph Bert, Luitpold Valentin Distel, Franziska Walter, Claus Belka, Sabine Semrau, Heinrich Iro, Rainer Fietkau, Yixing Huang, Florian Putz
Summary: This study introduces two novel methods for automatically analyzing the distribution of nodal metastases in head and neck cancer patients. The deep learning method accurately assigns lymph node metastases to different levels, while the registration-based method allows for the analysis and visualization of the 3D probability distribution of metastases without predefined level boundaries.
Article
Oncology
Luzie Helene Unseld, Laura S. Hildebrand, Florian Putz, Maike Buettner-Herold, Christoph Daniel, Rainer Fietkau, Luitpold Valentin Distel
Summary: Non-professional phagocytosis, specifically cell-in-cell structures, are increasingly studied in malignant melanoma metastasis. This study examined primary melanoma cell lines and tissue samples and found that all primary melanoma cell lines were capable of phagocytosis, with higher rates observed during suspension co-incubation. The presence of cell-in-cell structures was also detected in most tissue samples, indicating that non-professional phagocytosis mechanism is ubiquitous in malignant melanoma.