Three Dimensional Reconstruction Models for Medical Modalities: A Comprehensive Investigation and Analysis

Surgery of craniosynostosis: a historical review

Calvarial sutures and skull-shape deformities have been recognized since ancient times, but their direct link to premature suture fusion was first established in the 19th century. The earliest surgical attempts for craniosynostosis emerged in 1890 with strip craniectomy, though early outcomes were largely unsuccessful. Surgical progress stagnated for decades due to complications and skepticism surrounding these techniques. The 20th century brought significant advancements, beginning with successful strip craniectomies in the 1920s and later attempts to prevent re-ossification through material barriers. The late 1960s marked a pivotal shift with the introduction of cranial vault remodeling, followed by the integration of external cranial vault devices and helmet therapy in the 1980s. By the 1990s, endoscopic strip craniectomy revolutionized treatment by minimizing invasiveness and blood loss. Despite these innovations, debates persist regarding optimal surgical timing, long-term outcomes, and patient adherence to treatment. Emerging technologies such as 3D imaging, artificial intelligence, and personalized medicine hold promise for the future of craniosynostosis management.

Performance comparison in workflow efficiency between a remotely installed 3D workstation and an on-premises image processing workstation for dental cone-beam CT image reconstruction

OBJECTIVES

This study aims to compare the image processing times of dental cone beam CT (CBCT) images using a remote medical image processing workstation (RW) versus on-premises image processing (OP) and assess its impact on workflow efficiency.

METHODS

Data from 100 CBCT cases were randomly selected and processed using the OP3D VISION 17-19DX (EH Japan Co., Ltd.). In the OP environment, OnDemand 3D Dental (Cybermed Inc.) was used on a local terminal, while the RW setup involved a remote workstation-ZIO STATION (Ziosoft Inc.) connected via a 2 Gbps network. Seven experienced dentists processed the same data in both environments, and various processing times, including data transfer, re-slicing, 3D reconstruction, and PACS transfer, were compared.

RESULTS

The RW environment showed significantly shorter data transfer and re-slicing times than the OP environment. However, 3D image reconstruction times were similar between the two setups. Overall, processing time was significantly reduced in the RW environment. Variability in processing times among operators was observed, with most achieving reductions in the RW environment.

CONCLUSIONS

Remote processing of dental CBCT images using a dedicated image processing device offers equivalent or improved performance compared to on-premises processing. This approach can enhance workflow efficiency by reducing processing times and freeing up local resources, although further research is needed to optimize remote display protocols and multi-client access.

3D reconstruction of bone CT scan images based on deformable convex hull

Three-dimensional (3D) reconstruction of computed tomography (CT) and magnetic resonance imaging (MRI) images is an important diagnostic method, which is helpful for doctors to clearly recognize the 3D shape of the lesion and make the surgical plan. In the study of medical image reconstruction, most researchers use surface rendering or volume rendering method to construct 3D models from image sequences. The watertightness of the algorithm-reconstructed surface will be affected by the segmentation precision or the thickness of the CT layer. The articular surfaces at femoral ends are often used in biomechanical simulation experiments. The model may not conform to its original shape due to the manual repair of non-watertight surfaces. To solve this problem, a 3D reconstruction method of leg bones based on deep learning is proposed in this paper. By deforming the convex hull of the target, comparing with state-of-the-art methods, our method can stably generate a watertight model with higher reconstruction accuracy. In the situation of target transition structures getting fuzzy and the layer spacing increasing, the proposed method can maintain better reconstruction performance and appear higher robustness. Also, the chamfer loss is optimized based on the rotational shape of the leg bones, and the weight of the loss function can be assigned according to the geometric characteristics of the target. Experiment results show that the optimization method improves the accuracy of the model. Furthermore, our research provides a reference for the application of deep learning in medical image reconstruction.

Da Vinci Robot-Assisted Video Image Processing under Artificial Intelligence Vision Processing Technology

This research was aimed to explore the application value of intelligent algorithm-based digital images in Da Vinci robot-assisted treatment of patients with gastric cancer surgery. 154 patients were included as the research objects, with 89 cases in the control group underwent laparoscopic surgery, and 65 cases in the experimental group underwent robotic surgery. According to the propensity score, the patients in two groups were pair matched (1: 1), of which 104 cases (52 cases in each group) were successfully matched. The general data of patients, the changes in the images before and after the algorithm processing, the intraoperative and postoperative conditions, the pathological examination results, and the follow-up information were observed after matching. Compared with the original images, the images processed by the thread image edge detection algorithm had the significantly improved clarity, as well as highly reduced artifacts and noises. The sensitivity, specificity, and accuracy of image-assisted diagnosis were improved remarkably, showing the differences of statistical significance ( $P<0.05$ ). The total time of surgery, intraoperative bleeding, CRP (1d and 3d after surgery), and postoperative total abdominal drainage showed the significant differences as well ( $P<0.05$ ). The surgeries of patients in both groups met R0 resection (no tumor infiltration within 1 mm of the surgical margin), but there was a significant difference in the number of lymph node dissections ( $P<0.05$ ). The overall survival rates of patients in the experimental group and the control group were 83.0% and 76.1%, respectively, 2 years after surgery, with no significant difference ( $P>0.05$ ). The thread image edge detection algorithm produced a better processing effect on the images, which greatly improved the diagnostic sensitivity, specificity, and accuracy. Compared with endoscopic surgery, robotic surgery has better postoperative recovery, safety and reliability, and obvious advantages of minimally invasive surgery.

Extraction and Visualization of Ocular Blood Vessels in 3D Medical Images Based on Geometric Transformation Algorithm

Data extraction and visualization of 3D medical images of ocular blood vessels are performed by geometric transformation algorithm, which first performs random resonance response in a global sense to achieve detection of high-contrast coarse blood vessels and then redefines the input signal as a local image shielding the global detection result to achieve enhanced detection of low-contrast microfine vessels and complete multilevel random resonance segmentation detection. Finally, a random resonance detection method for fundus vessels based on scale decomposition is proposed, in which the images are scale decomposed, the high-frequency signals containing detailed information are randomly resonantly enhanced to achieve microfine vessel segmentation detection, and the final vessel segmentation detection results are obtained after fusing the low-frequency image signals. The optimal stochastic resonance response of the nonlinear model of neurons in the global sense is obtained to detect the high-grade intensity signal; then, the input signal is defined as a local image with high-contrast blood vessels removed, and the parameters are optimized before the detection of the low-grade intensity signal. Finally, the multilevel random resonance response is fused to obtain the segmentation results of the fundus retinal vessels. The sensitivity of the multilevel segmentation method proposed in this paper is significantly improved compared with the global random resonance results, indicating that the method proposed in this paper has obvious advantages in the segmentation of vessels with low-intensity levels. The image library was tested, and the experimental results showed that the new method has a better segmentation effect on low-contrast microscopic blood vessels. The new method not only makes full use of the noise for weak signal detection and segmentation but also provides a new idea of how to achieve multilevel segmentation and recognition of medical images.