RESUMO
BACKGROUND: Accurately identifying the branches of pulmonary segmental vessels and bronchi, as well as adjacent structures, and determining the spatial location of lesions within pulmonary segments, are major challenges for thoracic surgeons. The application of three-dimensional reconstruction technology holds promise in addressing this issue. OBJECTIVE: To evaluate the clinical value of three-dimensional reconstruction in thoracoscopic segmental surgery. METHODS: Seventy-seven patients who underwent thoracoscopic segmental surgery combined with three-dimensional reconstruction at our hospital from January 1, 2020, to August 31, 2023, were retrospectively analyzed. Preoperative chest enhanced CT scans were conducted, and MIMICS software aided in reconstructing DICOM format original data for patients with pulmonary nodules to facilitate intraoperative nodule localization. Accurate segmental pneumonectomy was performed by comparing preoperative anatomical identification of target segmental arteries, veins, and bronchi, with surgical details and postoperative outcomes recorded, including intraoperative pulmonary resection distribution, operation time, blood loss, chest tube drainage, extubation time, hospital stay, and complications. RESULTS: Following preoperative three-dimensional reconstruction, successful segmental lung surgeries were performed, predominantly with single segmental resection (92.2%), and a minority with combined segmentectomy (7.8%). Median operation time was 130225 minutes, with intraoperative blood loss at 70100 mL and postoperative chest tube drainage at 347 mL (159690 mL). Median extubation time and hospital stay were 4 days and 7 days, respectively. Complications within the 3-month follow-up affected 11.7% of cases, including persistent pulmonary leakage (7.1%), pulmonary infection (4.3%), atelectasis (4.3%), and pleural effusion (1.4%), with no fatalities. CONCLUSION: Preoperative 3D reconstruction can help the operator to perform safe, efficient and accurate thoracoscopic segmental pneumonectomy, which is worth popularizing in clinic.
RESUMO
It has not yet been proven whether sepsis affects the tissue around the anal canal. To address this issue, we established three-dimensional models for various types of anorectal abscesses and utilize 3D reconstruction of Magnetic Resonance Imaging scans to assess the extent of muscle damage caused by anorectal abscesses. Patients diagnosed with anorectal abscess, selected from January 2019 to January 2022 underwent pre- and post-operative scanning of pelvic floor and perianal tissues. The aforementioned structures were segmented for the reconstruction of a three-dimensional visual model and measurement of volumes for the abscess as well as the internal and external sphincters and levator ani muscle. The study included a total of 42 patients. Three-dimensional visualization models were created for different types of anorectal abscesses, including perianal, intersphincteric, ischiorectal, and supralevator abscesses. No statistically significant differences were observed in the volume of the internal sphincter, external sphincter, and levator ani muscle between pre- and post-operative patients. The 3D model of anorectal abscess, reconstructed from MRI data, offers a precise and direct visualization of the anatomical structures associated with various types of anorectal abscesses. The infection did not result in any damage to the internal and external anal sphincter and levator ani muscle.
Assuntos
Abscesso , Canal Anal , Imageamento Tridimensional , Imageamento por Ressonância Magnética , Humanos , Imageamento por Ressonância Magnética/métodos , Masculino , Feminino , Imageamento Tridimensional/métodos , Abscesso/diagnóstico por imagem , Abscesso/patologia , Pessoa de Meia-Idade , Adulto , Canal Anal/diagnóstico por imagem , Canal Anal/patologia , Idoso , Doenças do Ânus/diagnóstico por imagem , Doenças do Ânus/patologia , Doenças Retais/diagnóstico por imagem , Doenças Retais/patologia , Diafragma da Pelve/diagnóstico por imagem , Diafragma da Pelve/patologiaRESUMO
INTRODUCTION: Due to the complex anatomy of the spine and the intricate surgical procedures involved, spinal surgery demands a high level of technical expertise from surgeons. The clinical application of image-guided spinal surgery has significantly enhanced lesion visualization, reduced operation time, and improved surgical outcomes. AREAS COVERED: This article reviews the latest advancements in deep learning and artificial intelligence in image-guided spinal surgery, aiming to provide references and guidance for surgeons, engineers, and researchers involved in this field. EXPERT OPINION: Our analysis indicates that image-guided spinal surgery, augmented by artificial intelligence, outperforms traditional spinal surgery techniques. Moving forward, it is imperative to collect a more expansive dataset to further ensure the procedural safety of such surgeries. These insights carry significant implications for the integration of artificial intelligence in the medical field, ultimately poised to enhance the proficiency of surgeons and improve surgical outcomes.
RESUMO
OBJECTIVE: To compare the clinical value of multi-slice spiral computed tomography (MSCT) low-dose three-dimensional reconstruction and traditional X-ray in the auxiliary diagnosis of distal radius epiphyseal injury in children. METHODS: A retrospective analysis was performed on 105 children with distal radius bone scale injury (classified by Salter-Harris classification) admitted from March 2020 to June 2022. All children underwent MSCT three-dimensional reconstruction examination and traditional X-ray examination. The detection rate of epiphyseal injury of the distal radius was compared, along with the resolution, sensitivity and specificity. The image clarity and display degree of bone structure were analyzed. The radiation dose-related indicators and the time required for diagnosis were compared. RESULTS: The detection rate and diagnostic accuracy of MSCT (100%, 92.38%) was significantly higher than that of X-ray (76.19%, 64.76%). In terms of radiation dose index, the volume dose index CTDI of MSCT ranged from 1-5 mGy while the X-ray group ranged from 5-10 mGy. The dose length product (DLP) value of the MSCT group was lower than in the X-ray group (20-100 mGy·cm vs. 50-150 mGy·cm). The diagnostic scan time for MSCT was shorter than that of conventional X-ray. The acceptance rate with MSCT was 99%, significantly higher than that with conventional X-ray (85%). CONCLUSIONS: Low-dose three-dimensional reconstruction of MSCT in the diagnosis of epiphyseal injury of distal radius in children shows significant advantages over traditional CT in the detection rate, diagnostic accuracy, postoperative reduction quality evaluation, and radiation dose.
RESUMO
Background: Preoperative three-dimensional (3D) lung reconstructions can reduce intraoperative blood loss, conversion rate, and operation duration. These 3D reconstructions are predominantly provided by commercial expensive products, hence we aimed to assess the usability and performance of preoperative 3D lung reconstructions created with open-source software. Methods: Patients were invited to participate in this prospective pilot study if they were planned for uniportal video-assisted thoracoscopic surgery (VATS) lobectomy or segmentectomy between January and February 2023. Participants were excluded if a two-dimensional (2D) late-arterial-phase computed tomography (CT) scan contained motion artifacts, another surgical procedure was performed, or the surgery was canceled. After informed consent was obtained, 3D lung reconstructions were constructed using open-source 3D Slicer software. The system usability score (SUS) questionnaire assessed the usability of these reconstructions, whilst performance was evaluated based on anatomical validity compared to prior 2D CT assessment as well as operative findings. Descriptive statistics were reported. Results: Thirteen patients were included, of whom one underwent a segmentectomy. Eighty-three percent of the 3D lung reconstructions scored above average (SUS >68). Compared to 2D CT scans, 38% of lung nodule segmental locations were detected more accurately through 3D lung reconstructions. Furthermore, 3D lung reconstructions revealed anatomical variations in 62%, which were not recognized on 2D CT scans, and provided surgeons with insights that would change the procedure and/or transection planes in 62%. One 3D lung reconstruction failed to demonstrate an intraoperative recognized segmental pulmonary artery (A6) branch. Conclusions: Three-dimensional lung reconstructions created with open-source software were usable and effective for uniportal VATS anatomical resections. Trial Registration: ClinicalTrials.gov/NCT06132607.
RESUMO
BACKGROUND: Dermatofibrosarcoma protuberans (DFSP) is a superficial sarcoma characterized by infiltrative growth with tentacle-like borders. Mohs micrographic surgery (MMS) is the preferred treatment option for DFSP. However, the imprecise boundary localization in MMS leads to an increased number of Mohs layers required and a longer surgery time. High-frequency ultrasound has excellent tissue recognition capability for DFSP, allowing for precise boundary marking. MATERIALS AND METHODS: In this study, we retrospectively analyzed 14 cases of DFSP treated with MMS using preoperative ultrasound localization and three-dimensional reconstruction at Xiangya Hospital over the past 5 years. We also reviewed previous studies on MMS for DFSP treatment. RESULTS: It was found that the average number of Mohs layers for patients after preoperative ultrasound localization was 1.57, ranging from 1 to 3, which was less than the previously reported 1.86 layers, ranging from 1 to 12. This effectively reduced the number of Mohs layers required. CONCLUSIONS: By utilizing preoperative high-frequency ultrasound to determine the boundaries and depth of DFSP, the number of Mohs layers can be effectively reduced, leading to less workload for pathological examination, shorter operation time, and reduced surgical risks for patients. Ultrasound imaging data can be used for three-dimensional reconstruction, enabling less experienced Mohs surgeons to have a visual understanding of the morphology and extent of infiltration of the lesions. This aids in developing optimal surgical plans, smoothing the learning curve, and promoting the wider adoption of MMS.
RESUMO
INTRODUCTION AND HYPOTHESIS: The anatomy of the skeletal muscles located between the vagina and anus is important during complex obstetric laceration reconstructions. We aimed to clarify the composition of skeletal muscles located between the vagina and anal canal and their three-dimensional configuration relevant to perineum repair. METHODS: This observational study involved ten female cadavers. An anatomical dissection was performed to observe the muscles around the vagina and anal canal. Immunohistological analysis of the midsagittal section was performed to clarify the composition of the muscles, and dissection was performed to correspond to the cross-section. Wide-range serial sectioning and three-dimensional reconstruction were used to support these findings histologically and visualize the three-dimensional arrangement. RESULTS: The region between the vagina and anal canal included the anterior part of the external anal sphincter, superficial transverse perineal muscle approaching from the lateral side, and levator ani, located cranially. They converge three-dimensionally in the median from each direction, forming a muscle complex between the vagina and anal canal. CONCLUSIONS: The medial region between the vagina and anal canal in those giving birth includes a skeletal muscle complex formed by the confluence of the external anal sphincter, anterior bundle of the levator ani, and superficial transverse perineal muscle. In cases of severe perineal lacerations, these muscles could be injured. The anatomical knowledge that a part of the levator ani forms a muscle sling anterior to the anal canal is particularly important for obstetricians and gynecologists repairing obstetric lacerations and treating pelvic floor disorders.
RESUMO
The quality of underwater bridge piers significantly impacts bridge safety and long-term usability. To address limitations in conventional inspection methods, this paper presents a sonar-based technique for the three-dimensional (3D) reconstruction and visualization of underwater bridge piers. Advanced MS1000 scanning sonar is employed to detect and image bridge piers. Automated image preprocessing, including filtering, denoising, binarization, filling, and morphological operations, introduces an enhanced wavelet denoising method to accurately extract the foundation contour coordinates of bridge piers from sonar images. Using these coordinates, along with undamaged pier dimensions and sonar distances, a model-driven approach for a 3D pier reconstruction algorithm is developed. This algorithm leverages multiple sonar data points to reconstruct damaged piers through multiplication. The Visualization Toolkit (VTK) and surface contour methodology are utilized for 3D visualization, enabling interactive manipulation for enhanced observation and analysis. Experimental results indicate a relative error of 13.56% for the hole volume and 10.65% for the spalling volume, demonstrating accurate replication of bridge pier defect volumes by the reconstructed models. Experimental validation confirms the method's accuracy and effectiveness in reconstructing underwater bridge piers in three dimensions, providing robust support for safety assessments and contributing significantly to bridge stability and long-term safety assurance.
RESUMO
Early bifurcation of the common hepatic artery (EBCHA) is a rare anatomical variation (1%), that is often overlooked but can lead to accidental ligation of the right branch of the hepatic artery with consequent arterial ischemia of the right liver and potentially very serious complications during pancreaticoduodenectomy, partial hepatectomy, or liver harvesting for transplantation. It may be difficult to diagnose EBCHA using transverse imaging sections. However, on standard CT sections with intravenous contrast injection, three warning signs should allow the image reader to suspect it: presence of two hepatic arteries to the right of the celiac trunk, presence of a retro-portal hepatic artery, and absence of a right hepatic artery arising from the superior mesenteric artery. Analysis of the CT with reconstruction then allows for definitive diagnosis and limits the risk of accidental arterial injury or ligation.
Assuntos
Artéria Hepática , Humanos , Artéria Hepática/diagnóstico por imagem , Artéria Hepática/anormalidades , Tomografia Computadorizada por Raios X , Ligadura/métodos , Masculino , Feminino , Variação AnatômicaRESUMO
INTRODUCTION AND IMPORTANCE: We present a case of idiopathic azygos vein aneurysm (AVA) and describe its radiological features. Clinicians should be aware of the imaging findings to avoid the risk of bleeding caused by biopsy. CASE PRESENTATION: An asymptomatic 46-year-old woman was found to have an abnormal shadow on a chest computed tomography (CT) scan during a medical checkup. Plain CT revealed a well-defined, homogeneous mass on the posterior side of the right main bronchus. Contrast-enhanced CT revealed a mass with marked enhancement pointing to a vascular structure. Three-dimensional reconstruction showed that the mass was connected to the azygos arch, and inflow to and outflow from the azygos vein was detected. Subsequently, video-assisted thoracic surgery was performed. CLINICAL DISCUSSION: Patients with idiopathic AVA often present with accidental findings of a mediastinal or lung mass on a chest radiographs or CT scans, which can be mistakenly diagnosed as a paratracheal tumor, bronchial cyst, or posterior mediastinal tumor. Needle biopsy should be avoided due to the risk of massive bleeding. During the evaluation of thoracic malignancies, AVAs should be considered in the differential diagnosis, especially in area of the right tracheobronchial angle or right upper or posterior mediastinum. Three-dimensional reconstruction can help surgeons to clarify the disease diagnosis. CONCLUSION: Imaging findings combined with enhanced CT and three-dimensional reconstruction are useful for diagnosing AVAs.
RESUMO
OBJECTIVE: Segmentectomy has been widely performed in clinical practice, which required a comprehensive understanding of anatomical structure. In right lower lobe, studies of superior segment (S6) were relatively small-sized. And only one study focusing on subsuperior segment (S∗) was published, which showed different results with previous ones. As the close relationship between S6 and S∗, variation types and their prevalence rate were reported, aiming to providing larger-size study of S6 and showing new evidence on anatomical structure of S∗. METHODS: 800 CT imaging data were collected from patients in our hospital. Three-dimensional reconstruction was performed after quality check. All images were screened according to the definition of corresponding segment and anatomical variations were analyzed. RESULTS: The proportion of S6 variation types in the largest scale (718 patients) was reported with no significant difference compared to previous studies and newly classified subtypes of two-stem V6. The prevalence rate for S∗ in right lower lobe reached 28.3 % (203/718) with similar proportion of three types. Variation types and origins of pulmonary artery were analyzed in detail, finding two-stem A∗ only be observed in type III B∗. CONCLUSIONS: Through this study, the variation types and incidence rate of S6 were confirmed, and a different result of S∗ has been provided as well. The feasibility of the current classification standards and proposed new subclassifications were verified. The results would be a supplement to lung segmental anatomy and could advance researches in the future.
RESUMO
Despite recent advances in diagnosis and treatment, atherosclerotic coronary artery diseases remain a leading cause of death worldwide. Various imaging modalities and metrics can detect lesions and predict patients at risk; however, identifying unstable lesions is still difficult. Current techniques cannot fully capture the complex morphology-modulated mechanical responses that affect plaque stability, leading to catastrophic failure and mute the benefit of device and drug interventions. Finite Element (FE) simulations utilizing intravascular imaging OCT (Optical Coherence Tomography) are effective in defining physiological stress distributions. However, creating 3D FE simulations of coronary arteries from OCT images is challenging to fully automate given OCT frame sparsity, limited material contrast, and restricted penetration depth. To address such limitations, we developed an algorithmic approach to automatically produce 3D FE-ready digital twins from labeled OCT images. The 3D models are anatomically faithful and recapitulate mechanically relevant tissue lesion components, automatically producing morphologies structurally similar to manually constructed models whilst including more minute details. A mesh convergence study highlighted the ability to reach stress and strain convergence with average errors of just 5.9% and 1.6% respectively in comparison to FE models with approximately twice the number of elements in areas of refinement. Such an automated procedure will enable analysis of large clinical cohorts at a previously unattainable scale and opens the possibility for in-silico methods for patient specific diagnoses and treatment planning for coronary artery disease.
RESUMO
BACKGROUND: Endoscopic spine lumbar interbody fusion (Endo-LIF) is well-regarded within the academic community. However, it presents challenges such as intraoperative disorientation, high rates of nerve damage, a steep learning curve, and prolonged surgical times, often occurring during the creation of the operative channel. Furthermore, the undefined safe operational zones under endoscopy continue to pose risks to surgical safety. We aimed to analyse the anatomical data of Kambin's triangle via CT imaging to define the parameters of the safe operating area for transforaminal posterior lumbar interbody fusion (TPLIF), providing crucial insights for clinical practice. METHODS: We selected the L4-L5 intervertebral space. Using three-dimensional (3D), we identified Kambin's triangle and the endocircle within it, and recorded the position of point 'J' on the adjacent facet joint as the centre 'O' of the circle shifts by angle 'ß.' The diameter of the inscribed circle 'd,' the abduction angle 'ß,' and the distances 'L1' and 'L2' were measured from the trephine's edge to the exiting and traversing nerve roots, respectively. RESULTS: Using a trephine with a diameter of 8 mm in TPLIF has a significant safety distance. The safe operating area under the TPLIF microscope was also clarified. CONCLUSIONS: Through CT imaging research, combined with 3D simulation, we identified the anatomical data of the L4-L5 segment Kambin's triangle, to clarify the safe operation area under TPLIF. We propose a simple and easy positioning method and provide a novel surgical technique to establish working channels faster and reduce nerve damage rates. At the same time, according to this method, the Kambin's triangle anatomical data of the patient's lumbar spine diseased segments can be measured through CT 3D reconstruction of the lumbar spine, and individualised preoperative design can be conducted to select the appropriate specifications of visible trephine and supporting tools. This may effectively reduce the learning curve, shorten the time operation time, and improve surgical safety.
Assuntos
Imageamento Tridimensional , Vértebras Lombares , Fusão Vertebral , Tomografia Computadorizada por Raios X , Humanos , Fusão Vertebral/métodos , Vértebras Lombares/diagnóstico por imagem , Vértebras Lombares/cirurgia , Imageamento Tridimensional/métodos , Tomografia Computadorizada por Raios X/métodos , Masculino , Feminino , Pessoa de Meia-Idade , Endoscopia/métodos , Modelos Anatômicos , IdosoRESUMO
BACKGROUND: As a facial feature, the auricle plays an important role in the integrity and aesthetics of the whole face. Auricular subunits are associated with patient satisfaction in auricular reconstruction, but there are few studies on auricular subunits. We want to evaluate the reproducibility of auricular subunits by measuring the coordinates of the marker points of auricular subunits, accordingly provide a reference for the improvement of auricular reconstruction and the aesthetics of auricular injection. METHODS: Mimics 19.0 was used to carry out three-dimensional (3D) reconstruction of the computed tomography (CT) scan data of patients' brains; measure the three-dimensional coordinates of the 13 auricular subunit markers, the morphological auricle length and width, and the physiological auricle length and width; and analyze the reproducibility as well as the differences between group. RESULTS: Reproducibility of auricle subunit markers: There are 1124 (58.82%) high reproducibility, 580 (30.35%) moderate reproducibility, and 207 (10.83%) low reproducibility. The superior tragus notch, tragus, and antitragus had the highest reproducibility. There was no significant difference between the groups in the marking points on the helix, and there were no statistically significant differences in the measurement values of the auricles on the two sides. The physiological ear length and width and the morphological ear length of males were larger than those of females. These showed significant differences between the age groups. CONCLUSION: Most auricular subunit markers have high reproducibility. The subunits with higher reproducibility are the structures that need to be optimized during auricle reconstruction surgery or auricle injection in the future. NO LEVEL ASSIGNED: This journal requires that authors assign a level of evidence to each submission to which Evidence-Based Medicine rankings are applicable. This excludes Review Articles, Book Reviews, and manuscripts that concern Basic Science, Animal Studies, Cadaver Studies, and Experimental Studies. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .
RESUMO
We reported a case of heterotopic twin pregnancy in a unicornuate uterus with a non-communicating rudimentary horn with survival of both fetuses. The diagnosis was made late at 28 weeks of gestation, with suspicion raised by ultrasound and confirmed by magnetic resonance imaging (MRI). During hospitalization, obstetric ultrasound with color Doppler was performed every 2 days to assess fetal well-being and myometrial thickness, which was determined by measurements of the uterine wall at the accessory horn. Elective cesarean section was performed at 33 + 5 weeks of gestation. Delivery started with the fetus in the rudimentary horn, with subsequent extraction of the fetus in the unicornate uterus. Three-dimensional virtual reconstruction allowed a spatial view of the both uterus and fetuses with better understanding of the obstetrical condition by the parents and interactive discussion by the multidisciplinary medical team.
RESUMO
INTRODUCTION: In dental clinical practice, cone-beam computed tomography (CBCT) is commonly used to assist practitioners to recognize the complex morphology of root canal systems; however, because of its resolution limitations, certain small anatomical structures still cannot be accurately recognized on CBCT. The purpose of this study was to perform image super-resolution (SR) processing on CBCT images of extracted human teeth with the help of a deep learning model, and to compare the differences among CBCT, super-resolution computed tomography (SRCT), and micro-computed tomography (Micro-CT) images through three-dimensional reconstruction. METHODS: The deep learning model (Basicvsr++) was selected and modified. The dataset consisted of 171 extracted teeth that met inclusion criteria, with 40 maxillary first molars as the training set and 40 maxillary first molars as well as 91 teeth from other tooth positions as the external test set. The corresponding CBCT, SRCT, and Micro-CT images of each tooth in test sets were reconstructed using Mimics Research 17.0, and the root canal recognition rates in the 3 groups were recorded. The following parameters were measured: volume of hard tissue (V1), volume of pulp chamber and root canal system (V2), length of visible root canals under orifice (VL-X, where X represents the specific root canal), and intersection angle between coronal axis of canal and long axis of tooth (â X, where X represents the specific root canal). Data were statistically analyzed between CBCT and SRCT images using paired sample t-test and Wilcoxon test analysis, with the measurement from Micro-CT images as the gold standard. RESULTS: Images from all tested teeth were successfully processed with the SR program. In 4-canal maxillary first molar, identification of MB2 was 72% (18/25) in CBCT group, 92% (23/25) in SRCT group, and 100% (25/25) in Micro-CT group. The difference of hard tissue volume between SRCT and Micro-CT was significantly smaller than that between CBCT and Micro-CT in all tested teeth except 4-canal mandibular first molar (P < .05). Similar results were obtained in volume of pulp chamber and root canal system in all tested teeth (P < .05). As for length of visible root canals under orifice, the difference between SRCT and Micro-CT was significantly smaller than that between CBCT and Micro-CT (P < .05) in most root canals. CONCLUSIONS: The deep learning model developed in this study helps to optimize the root canal morphology of extracted teeth in CBCT. And it may be helpful for the identification of MB2 in the maxillary first molar.
RESUMO
The brain networks responsible for adaptive behavioral changes are based on the physical connections between neurons. Light and electron microscopy have long been used to study neural projections and the physical connections between neurons. Volume electron microscopy has recently expanded its scale of analysis due to methodological advances, resulting in complete wiring maps of neurites in a large volume of brain tissues and even entire nervous systems in a growing number of species. However, structural approaches frequently suffer from inherent limitations in which elements in images are identified solely by morphological criteria. Recently, an increasing number of tools and technologies have been developed to characterize cells and cellular components in the context of molecules and gene expression. These advancements include newly developed probes for visualization in electron microscopic images as well as correlative integration methods for the same elements across multiple microscopic modalities. Such approaches advance our understanding of interactions between specific neurons and circuits and may help to elucidate novel aspects of the basal ganglia network involving dopamine neurons. These advancements are expected to reveal mechanisms for processing adaptive changes in specific neural circuits that modulate brain functions.
RESUMO
OBJECTIVES: To develop a deep learning model for automated age estimation based on 3D CT reconstructed images of Han population in western China, and evaluate its feasibility and reliability. METHODS: The retrospective pelvic CT imaging data of 1 200 samples (600 males and 600 females) aged 20.0 to 80.0 years in western China were collected and reconstructed into 3D virtual bone models. The images of the ischial tuberosity feature region were extracted to create sex-specific and left/right site-specific sample libraries. Using the ResNet34 model, 500 samples of different sexes were randomly selected as training and verification set, the remaining samples were used as testing set. Initialization and transfer learning were used to train images that distinguish sex and left/right site. Mean absolute error (MAE) and root mean square error (RMSE) were used as primary indicators to evaluate the model. RESULTS: Prediction results varied between sexes, with bilateral models outperformed left/right unilateral ones, and transfer learning models showed superior performance over initial models. In the prediction results of bilateral transfer learning models, the male MAE was 7.74 years and RMSE was 9.73 years, the female MAE was 6.27 years and RMSE was 7.82 years, and the mixed sexes MAE was 6.64 years and RMSE was 8.43 years. CONCLUSIONS: The skeletal age estimation model, utilizing ischial tuberosity images of Han population in western China and employing the ResNet34 combined with transfer learning, can effectively estimate adult ischium age.
Assuntos
Determinação da Idade pelo Esqueleto , Aprendizado Profundo , Imageamento Tridimensional , Ísquio , Tomografia Computadorizada por Raios X , Humanos , Masculino , Feminino , Ísquio/diagnóstico por imagem , Adulto , Pessoa de Meia-Idade , Tomografia Computadorizada por Raios X/métodos , Imageamento Tridimensional/métodos , China , Estudos Retrospectivos , Determinação da Idade pelo Esqueleto/métodos , Idoso , Adulto Jovem , Idoso de 80 Anos ou mais , Reprodutibilidade dos TestesRESUMO
OBJECTIVE: This study investigates the clinical efficacy of integrating digital design with three-dimension (3D) printing technology in the transplantation of flaps for fingertip defects. METHODS: A retrospective analysis was conducted from October 2019 to June 2021 on 90 cases of patients with fingertip defects. These included 45 cases in which digital design, coupled with 3D printing, assisted the operation (3D printing group), and another 45 cases where patients underwent traditional pedicle flap transplantation and skin grafting (traditional operation group). A six-month postoperative follow-up assessed various measurements between the two groups, comparing the skin flap survival rate, aesthetic outcome, cold intolerance, sensory recovery, and overall skin flap performance. RESULTS: â Statistical analysis utilizing the independent samples t-test revealed a significant reduction in both operation time and flap anastomosis rate for the 3D printing group compared to the traditional operation group (P < 0.05). â¡ Conversely, the survival rate, aesthetic outcome, and cold intolerance showed no significant disparities between the groups (P > 0.05). ⢠Further, the Mann-Whitney U test indicated no significant difference in sensory recovery and overall efficacy assessment between the two cohorts (P > 0.05). CONCLUSION: Integrating digital design with 3D printing technology facilitated the surgical management of fingertip defects, achieving customized and precise approaches in flap transplantation. This precision in personalized skin flap design contributed to reduced operative time and enhanced surgical efficiency in such procedures.