Gender- and Age-Related Characterization of Lower Eyelid Morphology: Three-Dimensional Analysis in a Chinese Population.

BACKGROUND: The lower eyelid region is a critical component of the face. It is essential to establish anthropometric reference values for the evaluation of aging, surgical planning and assessment of outcomes in periocular esthetic and rejuvenation procedures. This study aims to provide comprehensive anthropometric data on the Chinese lower eyelid region, into account factors such as sex and age, through three-dimensional imaging analysis. METHOD: Three-dimensional facial images were obtained from 84 healthy Chinese individuals aged between 20-35 and 50-65 years, as well as eight patients aged between 20 and 35 who presented with eyelid bags. A total of 27 landmarks were identified, leading to the generation of corresponding 21 lines, 5 curves, 4 angles, 2 areas and 5 ratios. The measurements were compared among different age groups, genders and young patients with or without eyelid bags. RESULTS: Compared to females, males exhibited a more elongated palpebral fissure, lower tear trough and lid-cheek junction, smaller inner and outer canthus angles, as well as a larger area and proportion of the lower palpebral region. As age progressed, the height and width of the palpebral fissure and inner canthus angle decreased gradually, which was accompanied by sagging of the tear trough and lid-cheek junction, an increase in lower eyelid area and swelling of the lower eyelid. Young patients undergoing eyelid bags demonstrated larger and more swelling lower eyelid which held clinical significance for rejuvenation surgery. CONCLUSION: Males exhibited a higher proportion of the brow-eye unit occupied by the lower eyelid region compared to females. Elderly individuals displayed noticeable drooping of the tear trough and lid-cheek junction, accompanied by swelling in the lower palpebral region. These findings can serve as standard references for esthetic procedures and reconstructive periocular operations. LEVEL OF EVIDENCE IV: This journal requires that authors assign a level of evidence to each article. For a full description of these evidence-based medicine ratings, please refer to Table of Contents or online Instructions to Authors www.springer.com/00266.

A fully automatic MRI-guided decision support system for lumbar disc herniation using machine learning.

Background: Normalized decision support system for lumbar disc herniation (LDH) will improve reproducibility compared with subjective clinical diagnosis and treatment. Magnetic resonance imaging (MRI) plays an essential role in the evaluation of LDH. This study aimed to develop an MRI-based decision support system for LDH, which evaluates lumbar discs in a reproducible, consistent, and reliable manner. Methods: The research team proposed a system based on machine learning that was trained and tested by a large, manually labeled data set comprising 217 patients' MRI scans (3255 lumbar discs). The system analyzes the radiological features of identified discs to diagnose herniation and classifies discs by Pfirrmann grade and MSU classification. Based on the assessment, the system provides clinical advice. Results: Eventually, the accuracy of the diagnosis process reached 95.83%. An 83.5% agreement was observed between the system's prediction and the ground-truth in the Pfirrmann grade. In the case of MSU classification, 95.0% precision was achieved. With the assistance of this system, the accuracy, interpretation efficiency and interrater agreement among surgeons were improved substantially. Conclusion: This system showed considerable accuracy and efficiency, and therefore could serve as an objective reference for the diagnosis and treatment procedure in clinical practice.

Automated 3D Perioral Landmark Detection Using High-Resolution Network: Artificial Intelligence-Based Anthropometric Analysis.

BACKGROUND: 3D facial stereophotogrammetry, as a convenient, non-invasive and highly reliable evaluation tool, has shown great potential in pre-operative planning and treatment efficacy evaluation of plastic surgery in recent years. However, it requires manual identification of facial landmarks by trained evaluators to obtain anthropometric data, which consumes large amount of time and effort. Automatic 3D facial landmark localization may facilitate fast data acquisition and eliminate evaluator error. OBJECTIVES: In this paper, we propose a novel deep-learning method based on dimension-transformation and key-point detection for automated 3D perioral landmark annotation. METHODS: The 3D facial model is transformed into 2D images on which High-Resolution Network is implemented for key point detection. The 2D coordinates of key points are then mapped back to the 3D model using mathematical methods to obtain the 3D landmark coordinates. This program was trained with 120 facial models and validated in 50 facial models. RESULTS: Our approach achieved satisfactory accuracy of 1.30 ± 0.68 mm error in landmark detection with an average processing time of 5.2 ± 0.21 seconds per model. And subsequent analysis based on these landmarks showed an error of 0.87 ± 1.02 mm for linear measurements and 5.62 ± 6.61° for angular measurements. CONCLUSIONS: This automated 3D perioral landmarking method could serve as an effective tool that enables fast and accurate anthropometric analysis of lip morphology for plastic surgery and aesthetic procedures.

Automated anatomical landmark detection on 3D facial images using U-NET-based deep learning algorithm.

Background: Facial anthropometry based on 3-dimensional (3D) imaging technology, or 3D photogrammetry, has gained increasing popularity among surgeons. It outperforms direct measurement and 2-dimensional (2D) photogrammetry because of many advantages. However, a main limitation of 3D photogrammetry is the time-consuming process of manual landmark localization. To address this problem, this study developed a U-NET-based deep learning algorithm to enable automated and accurate anatomical landmark detection on 3D facial models. Methods: The main structure of the algorithm stacked 2 U-NETs. In each U-NET block, we used 3×3 convolution kernel and rectified linear unit (ReLU) as activation function. A total of 200 3D images of healthy cases, acromegaly patients, and localized scleroderma patients were captured by Vectra H1 handheld 3D camera and input for algorithm training. The algorithm was tested to detect 20 landmarks on 3D images. Percentage of correct key points (PCK) and normalized mean error (NME) were used to evaluate facial landmark detection accuracy. Results: Among healthy cases, the average NME was 1.4 mm. The PCK reached 90% when the threshold was set to the clinically acceptable limit of 2 mm. The average NME was 2.8 and 2.2 mm among acromegaly patients and localized scleroderma patients, respectively. Conclusions: This study developed a deep learning algorithm for automated facial landmark detection on 3D images. The algorithm was innovatively validated in 3 different groups of participants. It achieved accurate landmark detection and improved the efficiency of 3D image analysis.

Deep learning for diagnosing osteonecrosis of the femoral head based on magnetic resonance imaging.

BACKGROUND AND OBJECTIVE: Early-stage osteonecrosis of the femoral head (ONFH) can be difficult to detect because of a lack of symptoms. Magnetic resonance imaging (MRI) is sufficiently sensitive to detect ONFH; however, the diagnosis of ONFH requires experience and is time consuming. We developed a fully automatic deep learning model for detecting early-stage ONFH lesions on MRI. METHODS: This was a single-center retrospective study. Between January 2016 and December 2019, 298 patients underwent MRI and were diagnosed with ONFH. Of these patients, 110 with early-stage ONFH were included. Using a 7:3 ratio, we randomly divided them into training and testing datasets. All 3640 segments were delineated as the ground truth definition. The diagnostic performance of our model was analyzed using the receiver operating characteristic curve with the area under the receiver operating characteristic curve (AUC) and Hausdorff distance (HD). Differences in the area between the prediction and ground truth definition were assessed using the Pearson correlation and Bland-Altman plot. RESULTS: Our model's AUC was 0.97 with a mean sensitivity of 0.95 (0.95, 0.96) and specificity of 0.97 (0.96, 0.97). Our model's prediction had similar results with the ground truth definition with an average HD of 1.491 and correlation coefficient (r) of 0.84. The bias of the Bland-Altman analyses was 1.4 px (-117.7-120.5 px). CONCLUSIONS: Our model could detect early-stage ONFH lesions in less time than the experts. However, future multicenter studies with larger data are required to further verify and improve our model.