Feasibility of middle colic artery as a landmark for superior mesenteric artery - first approach in laparoscopic pancreatoduodenectomy: a prospective study.

BACKGROUND: SMA-first approach in pancreatoduodenectomy (PD) has been widely applied in open surgery as well as laparoscopy. Finding the superior mesenteric artery (SMA), inferior pancreatoduodenal artery (IPDA), first jejunal artery (J1A) has become a great challenge in laparoscopic PD (LPD). Meanwhile, exposing the midde colic artery (MCA) might be a feasible approach to determine SMA, IPDA, and J1A. Our study aims to find the anatomical correlation between MCA and SMA, IPDA, J1A, especially in SMA-first approach LPD from the left. METHODS: Uncontrolled clinical trial with 33 patients undergoing LPD had preoperative contrast abdominal CT scan to analyze the anatomical relevance between MCA and SMA, J1A, IPDA. The operation was performed starting with exposing MCA in advance to find SMA, J1A and IPDA. The data was analyzed by SPSS 25.0. RESULTS: 90.9% of MCA started at 12-3 o'clock from SMA, the mean distance from the SMA root to the MCA and J1A was 56.4 mm and 37.4 mm, respectively. The distance between SMA and J1A was 19 mm. 72.7% J1A started at 9-12 o'clock, 69.7% J1A and IPDA had a common trunk. 78.8% IPDA started at 3-6 o'clock. 100% of the cases had J1A controlled intraoperatively, 81.8% for IPDA when approached from the left, 3% had MCA injury. The mean time to approach from the left was 98 min, median blood loss was 100 ml. CONCLUSION: Exposing MCA first helps determine SMA, J1A and IPDA safely, efficiently and faciliates SMA-first approach LPD from the left and complete dissection of the mesopancreas and lymph nodes.

Transfer learning for anatomical structure segmentation in otorhinolaryngology microsurgery.

BACKGROUND: Reducing the annotation burden is an active and meaningful area of artificial intelligence (AI) research. METHODS: Multiple datasets for the segmentation of two landmarks were constructed based on 41 257 labelled images and 6 different microsurgical scenarios. These datasets were trained using the multi-stage transfer learning (TL) methodology. RESULTS: The multi-stage TL enhanced segmentation performance over baseline (mIOU 0.6892 vs. 0.8869). Besides, Convolutional Neural Networks (CNNs) achieved a robust performance (mIOU 0.8917 vs. 0.8603) even when the training dataset size was reduced from 90% (30 078 images) to 10% (3342 images). When directly applying the weight from one certain surgical scenario to recognise the same target in images of other scenarios without training, CNNs still obtained an optimal mIOU of 0.6190 ± 0.0789. CONCLUSIONS: Model performance can be improved with TL in datasets with reduced size and increased complexity. It is feasible for data-based domain adaptation among different microsurgical fields.

Stability between single and segmental maxillary osteotomies in bimaxillary surgery for cleft-associated class III deformity: a CBCT study.

OBJECTIVE: Segmental Le Fort I osteotomy through the cleft is a common strategy to narrow the alveolar cleft in adults. This study compared skeletal stability between single and segmental Le Fort I osteotomies in patients with unilateral cleft lip and palate (UCLP). MATERIALS AND METHODS: This retrospective analysis examined 45 adults with complete UCLP-associated class III deformities who underwent bimaxillary surgery with either single (n = 30) or segmental (n = 15) Le Fort I advancement. Cone beam computed tomography (CBCT) scans of the facial skeleton were acquired before surgery, 1-week postsurgery, and at follow-up. Measures of landmarks from the CBCT images for the two treatment groups were compared for translation (left/right, posterior/anterior, superior/inferior) and rotation (yaw, roll, pitch). RESULTS: Postsurgery, the downward movement of the maxilla was larger in the segmental group than the single group. At follow-up, the maxilla moved backward in both groups, and upward in the segmental group. The mandible moved forward and upward and rotated upward in both groups. The amount of upward movement and rotation was larger in the segmental group than the single group. CONCLUSIONS: Two years after bimaxillary surgery in patients with UCLP-associated class III deformity, greater relapse was found after segmental Le Fort I osteotomies in vertical translation of the maxilla and mandible, and pitch rotation of the mandible compared with single Le Fort I osteotomies. CLINICAL RELEVANCE: The vertical relapse of the maxilla was larger after segmental Le Fort I advancement compared with single Le Fort I advancement in clefts.

Site and severity of facial asymmetry after bimaxillary surgery for class III deformity: a case-control study.

OBJECTIVES: The study aimed (1) to evaluate the site and severity of facial asymmetry in Class III patients before and after bimaxillary surgery, and (2) to identify the influence of initial severity and positional jaw asymmetry on residual facial asymmetry. MATERIALS AND METHODS: Preoperative and postoperative cone-beam computed tomography of 65 patients with Class III facial asymmetry who underwent bimaxillary surgery were evaluated. Five midline and 14 paramedian facial soft tissue landmarks were identified to assess facial asymmetry. The outcomes were compared to a control group consisting of 30 age- and gender-matched Class I subjects. The postoperative positional jaw asymmetry (i.e., shift, roll, yaw) of each osteotomy segment (maxilla, mandible, chin, ramus) was also measured. RESULTS: Before surgery, the asymmetry was more severe at the chin, middle and lower contour. Bimaxillary surgery effectively corrected facial asymmetry, particularly in achieving normalization of chin deviation. However, significant asymmetry persisted postoperatively in the middle and lower contour (p < 0.001 and p < 0.01, respectively), which was affected by the positional ramus asymmetry in the roll and shift. CONCLUSIONS: Deviation of the chin, middle and lower contour contributed significantly to overall facial asymmetry in Class III asymmetry. Despite normalization of the chin deviation after bimaxillary surgery, asymmetry persisted at the middle and lower contour, primarily as the result of insufficient correction of the positional ramus asymmetry. CLINICAL RELEVANCE: Understanding the residual asymmetry after bimaxillary surgery is important for minimizing deviation and optimizing the surgical planning for its correction.

Gerdy's tubercle as a novel anatomical landmark for the proximal tibial cut in total knee arthroplasty.

OBJECTIVES: This study aimed to explore the use of Gerdy's tubercle (GT) as an innovative and dependable anatomical landmark for the proximal tibial cut in total knee arthroplasty (TKA) in cases with extensive knee degeneration. MATERIALS AND METHODS: One hundred dry tibia bones and 10 formalin-fixed cadaveric knee specimens of both sexes were examined. A Zimmer NexGen tibial cutting guide and a Mitutoyo digital caliper were utilized to align the guide with the tibia's mechanical axis. The procedure was replicated on cadaver knees using a standardized medial parapatellar arthrotomy approach. Measurements included the distance from GT superior border to the resection line and the length of the tibia. A radiological study involving magnetic resonance imaging examinations of 48 patients, which were evaluated focusing on the upper border of GT and the least degenerated segment of the posterolateral part of the lateral tibial condyle, was conducted. RESULTS: Anatomical measurements of GT and proximal tibial areas in 110 specimens showed slight but consistent variations with cadaver measurements. Magnetic resonance imaging analysis of 48 patients revealed notable sex differences in the distance between the superior border of GT and the tibia's posterolateral surface. There was also a significant negative correlation between the distance from GT to the posterolateral corner and cartilage thickness. CONCLUSION: Proper alignment in TKA is crucial for success, but identifying an extra-articular landmark for horizontal tibial resection remains challenging, particularly in severely arthritic knees. This study introduces GT as a novel anatomical landmark for TKA, offering a more reliable reference for achieving desired joint levels in knees with significant degenerative changes.

A novel approach to craniofacial analysis using automated 3D landmarking of the skull.

Automatic dense 3D surface registration is a powerful technique for comprehensive 3D shape analysis that has found a successful application in human craniofacial morphology research, particularly within the mandibular and cranial vault regions. However, a notable gap exists when exploring the frontal aspect of the human skull, largely due to the intricate and unique nature of its cranial anatomy. To better examine this region, this study introduces a simplified single-surface craniofacial bone mask comprising of 6707 quasi-landmarks, which can aid in the classification and quantification of variation over human facial bone surfaces. Automatic craniofacial bone phenotyping was conducted on a dataset of 31 skull scans obtained through cone-beam computed tomography (CBCT) imaging. The MeshMonk framework facilitated the non-rigid alignment of the constructed craniofacial bone mask with each individual target mesh. To gauge the accuracy and reliability of this automated process, 20 anatomical facial landmarks were manually placed three times by three independent observers on the same set of images. Intra- and inter-observer error assessments were performed using root mean square (RMS) distances, revealing consistently low scores. Subsequently, the corresponding automatic landmarks were computed and juxtaposed with the manually placed landmarks. The average Euclidean distance between these two landmark sets was 1.5 mm, while centroid sizes exhibited noteworthy similarity. Intraclass coefficients (ICC) demonstrated a high level of concordance (> 0.988), with automatic landmarking showing significantly lower errors and variation. These results underscore the utility of this newly developed single-surface craniofacial bone mask, in conjunction with the MeshMonk framework, as a highly accurate and reliable method for automated phenotyping of the facial region of human skulls from CBCT and CT imagery. This craniofacial template bone mask expansion of the MeshMonk toolbox not only enhances our capacity to study craniofacial bone variation but also holds significant potential for shedding light on the genetic, developmental, and evolutionary underpinnings of the overall human craniofacial structure.

LOW CERTAINTY OF EVIDENCE SUPPORTS THE APPLICATION OF (AI) FOR THE AUTOMATIC DETECTION OF CEPHALOMETRIC LANDMARKS WITH PROSPECTS FOR IMPROVEMENTS.

ARTICLE TITLE AND BIBLIOGRAPHIC INFORMATION: Artificial Intelligence for Detecting Cephalometric Landmarks: A Systematic Review and Meta-analysis. J Digit Imaging. 2023 Jun;36(3):1158-1179. doi:10.1007/s10278-022-00766-w. SOURCE OF FUNDING: The study was financed in part by the Coordenacao de Aperfeicoamentode Pessoal de Nivel Superior-Brazil (CAPES)-Finance Code 001. TYPE OF STUDY/DESIGN: Systematic review and meta-analysis.

Comparison of craniomaxillofacial morphology in children with unilateral cleft lip and palate treated with/without presurgical nasoalveolar molding - a retrospective study with a mean age of 5 years.

OBJECTIVES: To three-dimensionally assess differences in craniomaxillofacial skeletal development in patients with operated unilateral cleft lip and palate (UCLP) treated with/without presurgical nasoalveolar molding (PNAM) with a mean age of 5 years. MATERIALS AND METHODS: Cone-beam CT radiographs of 30 patients with UCLP who had undergone PNAM and 34 patients with UCLP who did not receive PNAM were analyzed. The data were stored in DICOM file format and were imported into the Dolphin Imaging program for 3D image reconstruction and landmark identification. 33 landmarks, 17 linear and three angular variables representing craniofacial morphology were analyzed and compared by using the Mann-Whitney U tests. RESULTS: The vast majority of linear variables and 3D coordinates of landmark points reflecting craniofacial skeletal symmetry were not significantly different between the two groups. In terms of craniofacial skeletal development, the PNAM group had a significantly smaller anterior nasal spine offset in the midsagittal plane and a greater maxillary length compared to the non-PNAM group. CONCLUSIONS: Evaluations performed in early childhood showed that treatment with/without PNAM in the neonatal period was not a major factor influencing craniomaxillofacial hard tissue development in patients with UCLP; moreover, PNAM treatment showed significant correction of skeletal deviation at the base of the nose. CLINICAL RELEVANCE: Follow-up in early childhood has shown that PNAM treatment administered during the neonatal stage does not impede maxillary development and has benefits in correcting nasal floor deviation. It is a viable option for improving nasal deformity in children with unilateral cleft lip and palate.

Automatic extraction of facial median sagittal plane for patients with asymmetry based on the EDMA alignment algorithm.

BACKGROUND: We aimed to establish a novel method for automatically constructing three-dimensional (3D) median sagittal plane (MSP) for mandibular deviation patients, which can increase the efficiency of aesthetic evaluating treatment progress. We developed a Euclidean weighted Procrustes analysis (EWPA) algorithm for extracting 3D facial MSP based on the Euclidean distance matrix analysis, automatically assigning weight to facial anatomical landmarks. METHODS: Forty patients with mandibular deviation were recruited, and the Procrustes analysis (PA) algorithm based on the original mirror alignment and EWPA algorithm developed in this study were used to construct the MSP of each facial model of the patient as experimental groups 1 and 2, respectively. The expert-defined regional iterative closest point algorithm was used to construct the MSP as the reference group. The angle errors of the two experimental groups were compared to those of the reference group to evaluate their clinical suitability. RESULTS: The angle errors of the MSP constructed by the two EWPA and PA algorithms for the 40 patients were 1.39 ± 0.85°, 1.39 ± 0.78°, and 1.91 ± 0.80°, respectively. The two EWPA algorithms performed best in patients with moderate facial asymmetry, and in patients with severe facial asymmetry, the angle error was below 2°, which was a significant improvement over the PA algorithm. CONCLUSIONS: The clinical application of the EWPA algorithm based on 3D facial morphological analysis for constructing a 3D facial MSP for patients with mandibular deviated facial asymmetry deformity showed a significant improvement over the conventional PA algorithm and achieved the effect of a dental clinical expert-level diagnostic strategy.

Determining the position of the lingula and the mandibular foramen using the antilingula in orthognathic surgery.

BACKGROUND: The antilingula located on the lateral surface of the mandibular ramus has been served as a surgical landmark for the mandibular foramen on the medial surface for decades. However, whether the antilingula truly represents the lingula which is the bony prominence overlapping the mandibular foramen, or the foramen itself, is still unclear. This study thus aimed to examine the position of the antilingula in relation to three reference points: the lingula, the anterior and the posterior borders of the mandibular foramen, as well as to the reference plane used in the inferior alveolar nerve block, and to the posterior border of the mandible. METHODS: This observational study was performed in 113 Thai dry mandibles. The antilingula were identified followed by transferring the reference points to the lateral surface. The distances from the antilingula to the reference points, the reference plane and the posterior border of the ramus were then measured. Chi-square test was calculated for side-dependency of the antilingula. Paired t-test was calculated for difference in measurements in left and right sides. RESULTS: The antilingula could be identified in 92.48% of the mandibles with 86.67 - 90.00% accuracy and 86.67% reliability. There was no significant difference in the presence of the antilingula on left and right sides (p = 0.801). Only 2.5% and 0.83% of the antilingula correspond to the lingula and the anterior border of the mandibular foramen, respectively. However, 85% of the reference points were located within 11 mm radius. The antilingula was found located 2.80 mm inferior to the reference plane and 16.84 mm from the posterior border of the ramus. CONCLUSIONS: The antilingula does not concur with the reference points on the medial surface. Our study also suggests that the safe area for vertical osteotomy is 11 mm posterior to the antilingula or at 30% of the length from the posterior border parallel to the occlusal plane. The use of more accurate techniques in localizing the mandibular foramen combined with the antilingula is more recommended than using the antilingula as a sole surgical guide.

Nerve Block Guided by Anatomic Landmarks Only Reduces Pain in Botulinum Toxin Type A Treatment for Glabellar and Forehead Wrinkles.

BACKGROUND: Botulinum toxin type A is widely used to treat glabellar and forehead wrinkles, but the pain caused by multiple injections often deters patients from receiving long-term treatment. Despite several methods used to alleviate this pain, consistency and effectiveness remain a challenge. Therefore, this study aimed to evaluate the effectiveness and safety of nerve block guided by anatomic landmarks only in reducing pain associated with botulinum toxin type A injections. PATIENTS AND METHODS: Between 2018 and 2022, the study enrolled 90 patients divided into 3 groups: the nerve block group (n = 30), the lidocaine cream group (n = 30), and the control group (n = 30). In the nerve block group, a landmarks-based technique was used to perform the nerve block. The study collected general information and comorbidities, and recorded pain at each point and time spent on preparation and treatment for each patient's forehead and glabellar area on each side. Patient-reported outcomes and complications were followed up at 2, 4, and 12 weeks after the injections. RESULTS: The nerve block group had significantly lower total pain scores in all regions compared to the lidocaine cream and control groups (P < 0.01). There were no significant differences in patient-reported outcomes between the groups at any follow-up point. Additionally, the complication rates related to injection were low and comparable among the 3 groups. CONCLUSIONS: Nerve block guided by anatomic landmarks only is a safe, effective, and consistent approach to reduce pain during botulinum toxin type A treatment for glabellar and forehead lines. This technique may offer advantages over other methods used to alleviate the pain associated with these injections.

Use of bone landmarks for assessing the safety of acupuncture on the posterior midline of the neck region.

OBJECTIVE: Many acupuncture acupoints are located on the posterior midline of the neck region. The needling depth for acupuncture is important for practitioners, and an unsafe needling depth increases the possibility of damage to the spinal cord and brainstem. Can the safety of acupuncture be assessed by examining bone structures? We focused on this aim to carry out this study. METHODS: The shortest distance from the posterior border of the foramen magnum to the line joining both upper ends of the posterior border of the mastoid process was measured on 29 skulls. Distances from the posterior border of the vertebral foramen to the tip of the spinous process and posterior tubercle of the atlas were measured and evaluated from 197 dry cervical vertebrae and 31 lateral cervical radiographs of patient subjects. The anatomic relationships of the vertebral canal with the external occipital protuberance, tip of the spinous process of the axis, tip of the posterior tubercle of the atlas, and upper end of the posterior border of the mastoid process were observed and evaluated via lateral cervical radiography. RESULTS: The shortest distance from the foramen magnum to the line between the mastoid processes was 4.65±1.75 mm, and the distance from the superior border of the vertebral foramen of the atlas to the posterior tubercle was less than the distance from the inferior border. The distance from the superior border of the vertebral canal to the tip of the spinous process in C2-C7 was greater than the distance from the inferior border. The mean lengths of the superior border of the C2 spinous process and the inferior border of the C7 spinous process were greater than 21 mm and 31 mm, respectively. The line from the upper end of the posterior border of the mastoid process to the tip of the C2 spinous process or 10 mm deep to the tip of the C2 spinous process was posterior to the vertebral canal. CONCLUSIONS: On the posterior midline of the neck region between the tip of spinous process of axis and external occipital protuberance, if the needle reaches the depth of the line between the upper end of posterior border of mastoid process and the tip of the spinous process of the axis, approximately 10 mm along the spinous process of the axis, the needle is in the safe region. The mean length of the C2-C7 spinous process is suitable to accommodate the needling depth of the adjacent acupoint. Bone structures can be used to effectively assess the safety of acupuncture on the posterior midline of the neck region.

Intraoperative trypan blue central landmark and its use in capsulotomy and capsulorhexis centration.

PURPOSE: To compare 3 capsulotomy centration methods. SETTING: Private clinic, Zlin, Czech Republic. DESIGN: Prospective, consecutive case series. METHODS: 180 eyes undergoing cataract surgery had anterior capsule staining with microfiltered 0.4% trypan blue solution before selective laser capsulotomy. The first 60 eyes (Group 1) had mydriatic dilated pupil centered capsulotomies. The next 60 eyes (Group 2) were centered on the trypan blue central landmark (TCL). The final 60 capsulotomies (Group 3) were centered on the patient fixated coaxial Purkinje reflex (CPR). Measurements between key anatomical landmarks and the TCL, CPR capsulotomies, and implanted intraocular lens (IOL) center were made. RESULTS: The TCL, observed in >94% of eyes in the study, coincided with the CPR with a displacement of <0.1 ± 0.1 mm. Group 1 capsulotomies were noticeably decentered on the IOLs by 0.3 ± 0.2 mm. The Group 2 symmetrical IOL relationship was maintained with a decentration of 0.15 ± 0.1 mm. Group 3 had a similar decentration with the IOLs with 0.15 ± 0.1 mm. Verification with IOLMaster 700 data and CALLISTO Eye System showed that the CPR and the TCL were coincident with the measured visual axis. CONCLUSIONS: The clearly visible TCL served as an alternate landmark to the patient fixated CPR, and being on the anterior capsule was not sensitive to tilt. Further patient compliance was not required. Both were superior to dilated pupil centration, to achieve symmetric IOL coverage. This has application for both capsulotomies and capsulorhexes.

Adrenal Vein Sampling: Tips and Tricks.

Adrenal vein sampling (AVS) is the standard method for distinguishing unilateral from bilateral sources of autonomous aldosterone production in patients with primary aldosteronism. This procedure has been performed at limited specialized centers due to its technical complexity. With recent advances in imaging technology and knowledge of adrenal vein anatomy in parallel with the development of adjunctive techniques, AVS has become easier to perform, even at nonspecialized centers. Although rare, anatomic variants of the adrenal veins can cause sampling failure or misinterpretation of the sampling results. The inferior accessory hepatic vein and the inferior emissary vein are useful anatomic landmarks for right adrenal vein cannulation, which is the most difficult and crucial step in AVS. Meticulous assessment of adrenal vein anatomy on multidetector CT images and the use of a catheter suitable for the anatomy are crucial for adrenal vein cannulation. Adjunctive techniques such as intraprocedural cortisol assay, cone-beam CT, and coaxial guidewire-catheter techniques are useful tools to confirm right adrenal vein cannulation or to troubleshoot difficult blood sampling. Interventional radiologists should be involved in interpreting the sampling results because technical factors may affect the results. In rare instances, bilateral adrenal suppression, in which aldosterone-to-cortisol ratios of both adrenal glands are lower than that of the inferior vena cava, can be encountered. Repeat sampling may be necessary in this situation. Collaboration with endocrinology and laboratory medicine services is of great importance to optimize the quality of the samples and for smooth and successful operation. ©RSNA, 2024 Test Your Knowledge questions for this article are available in the supplemental material.

Fenestration of the facial nerve by the stylomastoid artery.

BACKGROUND: Anatomic landmarks such as the tympanomastoid suture line, posterior belly of the digastric muscle, tragal pointer, and styloid process can assist the parotid surgeon in identifying and preserving the facial nerve. Vascular structures such as the posterior auricular artery and its branch, the stylomastoid artery, lay in close proximity to the facial nerve and have been proposed as landmarks for the identification of the facial nerve. In this case report, we describe an anatomic variation in which the stylomastoid artery has fenestrated the main trunk of the facial nerve, dividing it in two. METHODS: Two patients underwent parotidectomy (one for a pleomorphic adenoma, the second for a parotid cyst) through a standard anterograde approach with identification of the usual facial nerve landmarks. RESULTS: The appearance of the main trunk of the facial nerve was unusual in both patients due to its being fenestrated by the stylomastoid artery. The stylomastoid artery was divided, and the remainder of the facial nerve dissection was performed uneventfully with subsequent resection of the parotid mass in both patients. CONCLUSIONS: In rare instances, the stylomastoid artery can penetrate through the common trunk of the facial nerve. This is an important anatomic variant for the parotid surgeon to be aware of, as it can increase the difficulty of facial nerve dissection.

BM-BronchoLC - A rich bronchoscopy dataset for anatomical landmarks and lung cancer lesion recognition.

Flexible bronchoscopy has revolutionized respiratory disease diagnosis. It offers direct visualization and detection of airway abnormalities, including lung cancer lesions. Accurate identification of airway lesions during flexible bronchoscopy plays an important role in the lung cancer diagnosis. The application of artificial intelligence (AI) aims to support physicians in recognizing anatomical landmarks and lung cancer lesions within bronchoscopic imagery. This work described the development of BM-BronchoLC, a rich bronchoscopy dataset encompassing 106 lung cancer and 102 non-lung cancer patients. The dataset incorporates detailed localization and categorical annotations for both anatomical landmarks and lesions, meticulously conducted by senior doctors at Bach Mai Hospital, Vietnam. To assess the dataset's quality, we evaluate two prevalent AI backbone models, namely UNet++ and ESFPNet, on the image segmentation and classification tasks with single-task and multi-task learning paradigms. We present BM-BronchoLC as a reference dataset in developing AI models to assist diagnostic accuracy for anatomical landmarks and lung cancer lesions in bronchoscopy data.

Deep learning for automatic detection of cephalometric landmarks on lateral cephalometric radiographs using the Mask Region-based Convolutional Neural Network: a pilot study.

OBJECTIVE: We examined the effectiveness and feasibility of the Mask Region-based Convolutional Neural Network (Mask R-CNN) for automatic detection of cephalometric landmarks on lateral cephalometric radiographs (LCRs). STUDY DESIGN: In total, 400 LCRs, each with 19 manually identified landmarks, were collected. Of this total, 320 images were randomly selected as the training dataset for Mask R-CNN, and the remaining 80 images were used for testing the automatic detection of the 19 cephalometric landmarks, for a total of 1520 landmarks. Detection rate, average error, and detection accuracy rate were calculated to assess Mask R-CNN performance. RESULTS: Of the 1520 landmarks, 1494 were detected, for a detection rate of 98.29%. The average error, or linear deviation distance between the detected points and the originally marked points of each detected landmark, ranged from 0.56 to 9.51 mm, with an average of 2.19 mm. For detection accuracy rate, 649 landmarks (43.44%) had a linear deviation distance less than 1 mm, 1020 (68.27%) less than 2 mm, and 1281 (85.74%) less than 4 mm in deviation from the manually marked point. The average detection time was 1.48 seconds per image. CONCLUSIONS: Deep learning Mask R-CNN shows promise in enhancing cephalometric analysis by automating landmark detection on LCRs, addressing the limitations of manual analysis, and demonstrating effectiveness and feasibility.

Fully automated landmarking and facial segmentation on 3D photographs.

Three-dimensional facial stereophotogrammetry provides a detailed representation of craniofacial soft tissue without the use of ionizing radiation. While manual annotation of landmarks serves as the current gold standard for cephalometric analysis, it is a time-consuming process and is prone to human error. The aim in this study was to develop and evaluate an automated cephalometric annotation method using a deep learning-based approach. Ten landmarks were manually annotated on 2897 3D facial photographs. The automated landmarking workflow involved two successive DiffusionNet models. The dataset was randomly divided into a training and test dataset. The precision of the workflow was evaluated by calculating the Euclidean distances between the automated and manual landmarks and compared to the intra-observer and inter-observer variability of manual annotation and a semi-automated landmarking method. The workflow was successful in 98.6% of all test cases. The deep learning-based landmarking method achieved precise and consistent landmark annotation. The mean precision of 1.69 ± 1.15 mm was comparable to the inter-observer variability (1.31 ± 0.91 mm) of manual annotation. Automated landmark annotation on 3D photographs was achieved with the DiffusionNet-based approach. The proposed method allows quantitative analysis of large datasets and may be used in diagnosis, follow-up, and virtual surgical planning.

Measurements of oculo-palpebral landmarks and evaluation of patient's head position.

PURPOSE: Keeping the head in a neutral position is requisite for glasses/lenses/head-up designs, the suitability of oculo-plastic surgery and for the grading the eye shift. Anatomically incompatible glasses are one of the common problems affecting accommodation, reducing comfort and disturbing by causing symptoms such as dizziness and nausea. The oculo-palpebral measurements act as a key determinant in symmetrical facial attractiveness. This study aims to investigate the most effective oculo-palpebral landmarks, head-neutral as the ideal position, taking into account of individual anatomical differences of these patients. METHODS: 100 females and 100 males aged between 18 and 20 years were photographed. Digital photogrammetric measurements were made with the ImageJ program. Interpupillary and interhelical distances, besides bilateral palpebral fissure length and height, and iris diameter were calculated on front-facing photographs. RESULTS: Mean interpupillary distance was measured wider in males than in females. The mean length of palpebral fissure was 31 mm; palpebral fissure height was 10 mm. These figures were valid in both eyes and gender. The interhelical distance was calculated as the mean and was measured longer in men. Since the measurement values were the same in both sexes and on both sides, they were determined as important landmarks for controlling the head-neutral position, evaluating whether there was a deviation in the eye, and measuring the numerical value when detected. CONCLUSION: It is essential to check the side-symmetry of the patient's palpebral fissure height, palpebral fissure length, diameter of iris and corneal depth during oculo-plastic invention and artificial design.