Clinical efficacy and performance evaluation of a bendable remote robot system for a bone tumour surgery: A pilot animal study.

BACKGROUND: Traditional open surgery for bone tumours sometimes has as a consequence an excessive removal of healthy bone tissue because of the limitations of rigid surgical instruments, increasing infection risk and recovery time. METHODS: We propose a remote robot with a 4.5-mm diameter bendable end-effector, offering four degrees of freedom for accessing the inside of the bone and performing tumour debridement. The preclinical studies evaluated the effectiveness, clinical scenario, and usability across 12 total surgeries-six phantom surgeries and six bovine bone surgeries. Evaluation criteria included skin incision size, bone window size, surgical time, removal rate, and conversion to open surgery. RESULTS: Preclinical studies demonstrated that the robotic approach requires significantly smaller incision size and procedure times than traditional open curettage. CONCLUSION: This study validated the performance of the proposed system by assessing its preclinical effectiveness and optimising surgical methods using human phantom and bovine bone tumour models.

Clinical evaluation of a fluoroscopic image-based laser guidance system in bone tumor surgery: A technical note.

This study presents a laser guidance system developed to enhance surgical accuracy and reduce radiation exposure in orthopedic surgeries. The system can project the actual position corresponding to the appointed position selected by the surgeon on a fluoroscopic image using a line laser and has laser projection ability to mark the corresponding point using a line laser. The surgeon does not have to perform anatomical marker placement for calibration. Three patients with bone tumors underwent surgeries using the laser guidance system, and the projection accuracy was evaluated by measuring the distance error between the appointed and laser-marking positions. The installation time, including calibration, was also assessed for clinical usability. The average projection accuracy in bone tumor surgery was 2.86 mm, and the average installation time was 7 min. These results demonstrate that the laser guidance system, with a projection error of <3 mm, could be useful in bone tumor surgeries.

Generative approach for data augmentation for deep learning-based bone surface segmentation from ultrasound images.

PURPOSE: Precise localization of cystic bone lesions is crucial for osteolytic bone tumor surgery. Recently, there is a move toward ultrasound imaging over plain radiographs (X-rays) for intra-operative navigation due to the radiation-free and cost-effectiveness of the modality. In this process, the intra-operative bone model reconstructed from the segmented ultrasound image is registered with the pre-operative bone model. Deep learning approaches have recently shown remarkable success in bone surface segmentation from ultrasound images. However, to train deep learning models effectively with limited dataset size, data augmentation is essential. This study investigates the applicability of the generative approach for data augmentation as well as identifies standard data augmentation approaches for bone surface segmentation from ultrasound images. METHODS: The generative approach we used in our work is based on Pix2Pix image-to-image translation network. We have proposed a multiple-snapshot approach, which mitigates the uni-modal deterministic output issue in the Pix2Pix network without using any complex architecture and training process. We also identified standard data augmentation approaches necessary for ultrasound bone surface segmentation through experiments. RESULTS: We have evaluated our networks using 800 ultrasound images from trained regions (humerus bone) and 1200 ultrasound images from untrained regions (tibia and femur bones) using four different augmentation approaches. The results show that the generative augmentation approach has a positive impact on accuracy in both trained (+ 4.88%) and untrained regions (+ 25.84%) compared to using only standard augmentations. Moreover, compared to standard augmentation approaches, the addition of the generative augmentation approach also showed a similar trend in both trained (+ 8.74%) and untrained (+ 11.55%) regions. CONCLUSION: Generative approaches are very beneficial for data augmentation, where limited dataset size is prevalent, such as ultrasound bone segmentation. The proposed multiple-snapshot Pix2Pix approach has the potential to generate multimodal images, which enlarges the dataset considerably.