3D Head Scans Research Articles

Ordering Eyeglasses Using 3D Head Scan Technology versus Established Online and Storefront Clinic Methods.

This study investigates how a new smartphone scanning technology compares with established online and storefront vendors in providing remote measurement and adjustment of prescription eyeglasses. This study aimed to evaluate a new technology for ordering prescription eyeglasses online. Thirty participants with 2.00 to 2.75 D of presbyopia (aged 49 to 74 years) were asked to order eyeglasses with progressive addition lenses from four vendors: one online vendor using a new head scan technology (Topology, San Francisco, CA), two established Web site-based vendors (vendors A and B), and one storefront vendor: Université de Montréal's Vision Clinic (UMVC). The resulting measurements were compared with those of opticians. Participant-reported preferences on visual and physical comfort of eyeglasses were collected after 15-minute trials of eyeglasses from each vendor. Pupillary half-distance measured with Topology matched optician measurements, but online vendors A and B diverged (mean difference, - 0.80 mm [ Z = -2.79; P = .005]). Topology and vendor B segment addition heights were similar to optician measurements, but vendor A diverged (mean, -1.40 mm [ Z = -2.58; P = .01]). The personalization parameter values obtained with Topology were different from optician measurements for pantoscopic angle (-5.30° [ Z = -4.12; P < .001]) and face wrap angle (+1.25° [ Z = -2.94; P = .003]). The UMVC eyeglasses scored best for adjustment (8.71/10 [ Z = -5.53; P < .001]), with Topology coming second (7.23/10). Topology scores were equivalent to UMVC scores for all eight items of patient-reported preferences (nonparametric Friedman analysis of variance, P < .05). Basic lens centration measurements obtained with Topology compare well with those of opticians, but some aspects of the methodology for measuring personalization parameters could be improved. In comparison with two established online vendors, resulting measurements with Topology are more consistent. Initial wearer satisfaction with Topology eyeglasses was also better.

Age and sex factors integrated 3D statistical models of adults' heads

3D parametric head models and 3D anthropometric analyses are essential in designing head-related products. Existing 3D parametric head modeling only analyzes sex-based dimension and shape variances but ignores age-related effects. Thus, we developed detailed 3D statistical models of adults' heads based on 1496 3D head scans of Chinese individuals aged 18 to 84, including a global model and four subclass models. An age threshold identification method has been presented to cluster the 3D heads into four groups based on their anthropometric features. The total explained variance of the first five principal components (PCs) in all head models was more than 76%. The relationships between the PCs, the head areas, and the specific dimensions of head regions have also been identified in this study. These models can be reliably used for fit assessment, mass customization, and virtual fittings for head-related products.

Identification of Extracapsular Extension in Head and Neck Cancer Patients Using Artificial Intelligence.

<h3>Purpose/Objective(s)</h3> Routine diagnostic CT scan guides the management of head and neck squamous cell carcinoma (HNSCC)) by identifying tumor and lymph node features. Nevertheless, lymph node extracapsular extension (ECE) remains challenging to detect by clinicians even using modern imaging techniques. It is essential to detect the occurrence of ECE as it changes staging and management for the patients. Current clinical ECE detection relies on clinical examination, radiological identification, and histopathological confirmation. Therefore, we aim to perform ECE identification automatically using a gradient-based class activation mapping (Grad-CAM) technique to guide the deep learning model to focus on the regions highly suspicious of ECE. The presence or absence of ECE has been analyzed and correlated with gold standard histopathological findings. <h3>Materials/Methods</h3> This research proposes a novel deep learning framework to detect ECE from 3D head and neck CT scans. The research problem is treated as a classification task, i.e., to label patients' CT scans with ECE positive or negative categories. Grad-CAM can provide visual explanations for deep neural networks by indicating the critical regions in the image for prediction. The proposed network has two convolutional neural networks (CNNs) involved. The first CNN model acts as a preprocessing step. The Grad-CAM method is applied to the trained CNN to highlight and extract the volumes of interest (VOIs) related to the class ECE positive. The second CNN model takes the extracted VOIs as model input for ECE positive/negative classification. The data used for the second CNN training does not involve that in the first step training process. The studied dataset includes 130 patients' CT scans. Cross-validation has been applied in our experimental tests. <h3>Results</h3> The experimental results have demonstrated that our model can identify ECE positive and negative patients based on ten-fold cross-validation. We have achieved 95.03% test accuracy and 90.90% test AUC in ECE classification. Furthermore, ECE probability heatmaps are generated by Grad-CAM. <h3>Conclusion</h3> This study demonstrates that a deep learning model guided by the Grad-CAM approach can be used for ECE detection. The proposed two-step neural networks can help capture sufficient features and achieve high classification accuracy. The heatmaps can be used as a tool for radiologists to assist in the identification of ECE. The outcome of this research is expected to promote the implementation of artificial intelligence for head and neck cancer ECE identification diagnosis in the radiology computer vision field.

Laplace-Beltrami Refined Shape Regression Applied to Neck Reconstruction for Craniosynostosis Patients

Abstract This contribution is part of a project concerning the creation of an artificial dataset comprising 3D head scans of craniosynostosis patients for a deep-learning-based classification. To conform to real data, both head and neck are required in the 3D scans. However, during patient recording, the neck is often covered by medical staff. Simply pasting an arbitrary neck leaves large gaps in the 3D mesh. We therefore use a publicly available statistical shape model (SSM) for neck reconstruction. However, most SSMs of the head are constructed using healthy subjects, so the full head reconstruction loses the craniosynostosis-specific head shape. We propose a method to recover the neck while keeping the pathological head shape intact. We propose a Laplace- Beltrami-based refinement step to deform the posterior mean shape of the full head model towards the pathological head. The artificial neck is created using the publicly available Liverpool-York-Model. We apply our method to construct artificial necks for head scans of 50 scaphocephaly patients. Our method reduces mean vertex correspondence error by approximately 1.3 mm compared to the ordinary posterior mean shape, preserves the pathological head shape, and creates a continuous transition between neck and head. The presented method showed good results for reconstructing a plausible neck to craniosynostosis patients. Easily generalized it might also be applicable to other pathological shapes.

Development of prototype hat patterns for elderly women based on three-dimensional modeling

Population aging is a global phenomenon, and the elderly population has a higher economic capability today than that in the past. Thus, this population is considered to be a growing consumer group that enjoys both consumer and leisure life. In this study, we developed prototype hat patterns for elderly women that can be used for developing close-fitting hats, helmets, masks, and smart headwear. Three-dimensional (3D) head scan data of elderly women were employed herein, and the data were classified into three groups with common head size and head shape. The target group was selected from a high-frequency group among the classified groups, and a standard head form representing the target group was developed using averaging and wire frame generation techniques. Four hat types were considered, design baselines were developed for each type, and prototype patterns were designed for each hat type using a flattening technique. The suitability of the developed prototype hat patterns was subsequently verified. Our results showed that all four prototype patterns had errors less than 5 mm2 (1.40%). The hat patterning method proposed in this study is expected to improve the wearing comfort of high-value-added products designed for the elderly.

The need for overcorrection: evaluation of computer-assisted, virtually planned, fronto-orbital advancement using postoperative 3D photography

The main indication for craniofacial remodeling of craniosynostosis is to correct the deformity, but potential increased intracranial pressure resulting in neurocognitive damage and neuropsychological disadvantages cannot be neglected. The relapse rate after fronto-orbital advancement (FOA) seems to be high; however, to date, objective measurement techniques do not exist. The aim of this study was to quantify the outcome of FOA using computer-assisted design (CAD) and computer-assisted manufacturing (CAM) to create individualized 3D-printed templates for correction of craniosynostosis, using postoperative 3D photographic head and face surface scans during follow-up. The authors included all patients who underwent FOA between 2014 and 2020 with individualized, CAD/CAM-based, 3D-printed templates and received postoperative 3D photographic face and head scans at follow-up. Since 2016, the authors have routinely planned an additional "overcorrection" of 3 mm to the CAD-based FOA correction of the affected side(s). The virtually planned supraorbital angle for FOA correction was compared with the postoperative supraorbital angle measured on postoperative 3D photographic head and face surface scans. The primary outcome was the delta between the planned CAD/CAM FOA correction and that achieved based on 3D photographs. Secondary outcomes included outcomes with and those without "overcorrection," time of surgery, blood loss, and morbidity. Short-term follow-up (mean 9 months after surgery; 14 patients) showed a delta of 12° between the planned and achieved supraorbital angle. Long-term follow-up (mean 23 months; 8 patients) showed stagnant supraorbital angles without a significant increase in relapse. Postsurgical supraorbital angles after an additionally planned overcorrection (of 3 mm) of the affected side showed a mean delta of 11° versus 14° without overcorrection. The perioperative and postoperative complication rates of the whole cohort (n = 36) were very low, and the mean (SD) intraoperative blood loss was 128 (60) ml with a mean (SD) transfused red blood cell volume of 133 (67) ml. Postoperative measurement of the applied FOA on 3D photographs is a feasible and objective method for assessment of surgical results. The delta between the FOA correction planned with CAD/CAM and the achieved correction can be analyzed on postoperative 3D photographs. In the future, calculation of the amount of "overcorrection" needed to avoid relapse of the affected side(s) after FOA may be possible with the aid of these techniques.

Nonrigid 3D motion estimation at high temporal resolution from prospectively undersampled k-space data using low-rank MR-MOTUS.

PurposeWith the recent introduction of the MR‐LINAC, an MR‐scanner combined with a radiotherapy LINAC, MR‐based motion estimation has become of increasing interest to (retrospectively) characterize tumor and organs‐at‐risk motion during radiotherapy. To this extent, we introduce low‐rank MR‐MOTUS, a framework to retrospectively reconstruct time‐resolved nonrigid 3D+t motion fields from a single low‐resolution reference image and prospectively undersampled k‐space data acquired during motion.TheoryLow‐rank MR‐MOTUS exploits spatiotemporal correlations in internal body motion with a low‐rank motion model, and inverts a signal model that relates motion fields directly to a reference image and k‐space data. The low‐rank model reduces the degrees‐of‐freedom, memory consumption, and reconstruction times by assuming a factorization of space‐time motion fields in spatial and temporal components.MethodsLow‐rank MR‐MOTUS was employed to estimate motion in 2D/3D abdominothoracic scans and 3D head scans. Data were acquired using golden‐ratio radial readouts. Reconstructed 2D and 3D respiratory motion fields were, respectively, validated against time‐resolved and respiratory‐resolved image reconstructions, and the head motion against static image reconstructions from fully sampled data acquired right before and right after the motion.ResultsResults show that 2D+t respiratory motion can be estimated retrospectively at 40.8 motion fields per second, 3D+t respiratory motion at 7.6 motion fields per second and 3D+t head‐neck motion at 9.3 motion fields per second. The validations show good consistency with image reconstructions.ConclusionsThe proposed framework can estimate time‐resolved nonrigid 3D motion fields, which allows to characterize drifts and intra and inter‐cycle patterns in breathing motion during radiotherapy, and could form the basis for real‐time MR‐guided radiotherapy.

Three-Dimensional Calvarial Growth in Spring-Assisted Cranioplasty for Correction of Sagittal Synostosis.

Spring-assisted cranioplasty (SAC) is a minimally invasive technique for treating sagittal synostosis in young infants. Yet, follow-up data on cranial growth in patients who have undergone SAC are lacking. This project aimed to understand how the cranial shape develops during the postoperative period, from spring insertion to removal. 3D head scans of 30 consecutive infants undergoing SAC for sagittal synostosis were acquired using a handheld scanner pre-operatively, immediately postoperatively, at follow-up and at spring removal; 3D scans of 41 age-matched control subjects were also acquired. Measurements of head length, width, height, circumference, and volume were taken for all subjects; cephalic index (CI) was calculated. Statistical shape modeling was used to compute 3D average head models of sagittal patients at the different time points. SAC was performed at a mean age of 5.2 months (range 3.3-8.0) and springs were removed 4.3 months later. CI increased significantly (P < 0.001) from pre-op (69.5% ± 2.8%) to spring removal (74.4% ± 3.9%), mainly due to the widening of head width, which became as wide as for age-matched controls; however, the CI of controls was not reached (82.3% ± 6.8%). The springs did not constrain volume changes and allowed for natural growth. Population mean shapes showed that the bony prominences seen at the sites of spring engagement settle over time, and that springs affect the overall 3D head shape of the skull. In conclusion, results reaffirmed the effectiveness of SAC as a treatment method for nonsyndromic single suture sagittal synostosis.

Passenger head kinematics in abrupt braking and lane change events

Objective: A test track study was conducted to quantify patterns of adult front seat passenger head motion during abrupt vehicle maneuvers.Method: Eighty-seven men and women with a wide range of body sizes and ages participated in data collection on a closed test track in a passenger sedan under manual control by a test driver. Because a primary goal of the study was to gather “unaware” data, the participants were instructed that the study was concerned with vehicle dynamics and they were required to read from a questionnaire taped to the top of their thighs as the drive began. The first event was a hard brake (approximately 1 g) to a stop from 35 mph (56 kph). Within the space of approximately 5 min the participants also experienced an aggressive lane change, a sharp right turn with simultaneous hard braking, and a second hard braking event. A Microsoft Kinect v2 sensor was positioned to view the area around the front passenger seat. Head location was tracked using the Kinect data with a novel methodology that fit 3D head scan data to the depth data acquired in the vehicle.Result: The mean (standard deviation) forward excursion of the estimated head center of gravity (CG) location in the first braking event was 135 (62) mm. The forward head CG excursion in the second braking event of 115 (51) mm was significantly less than that in the first, but the difference was small relative to the within-condition variance. Head excursion on the second braking trial was less than that on the first trial for 69% of participants. The mean maximum inboard head excursion in lane-change maneuvers was 118 (40) mm. Forward head excursions in braking were significantly smaller for older passengers and those with higher body mass index, but the combined factors accounted for less than 25% of the variance. Inboard head excursion in the lane-change event was significantly related to stature, but only about 7% of variance was related to body size. Head excursions for men and women did not differ significantly after accounting for body size.Discussion: This is the first quantitative occupant dynamics study to use a large, diverse sample of passengers, enabling the exploration of the effects of covariates such as age and body size.Conclusions: The data demonstrate that a relatively large range of head positions can be expected to result from abrupt vehicle maneuvers. The data do not support simple scaling of excursions based on body size.

A novel hierarchical clustering algorithm for the analysis of 3D anthropometric data of the human head

In recent years, the use of 3D anthropometry for product design has become more appealing because of advances in mesh parameterisation, multivariate analyses and clustering algorithms. The purpose of this study was to introduce a new method for the clustering of 3D head scans. A novel hierarchical algorithm was developed, in which a squared Euclidean metric was used to assess the head shape similarity of participants. A linkage criterion based on the centroid distance was implemented, while clusters were created one after another in an enhanced manner. As a result, 95.0% of the studied sample was classified inside one of the four computed clusters. Compared to conventional hierarchical techniques, our method could classify a higher ratio of individuals into a smaller number of clusters, while still satisfying the same variation requirements within each cluster. The proposed method can provide meaningful information about the head shape variation within a population, and should encourage ergonomists to use 3D anthropometric data during the design process of head and facial gear.

Cranial bone structure in children with sagittal craniosynostosis: Relationship with surgical outcomes

While spring-assisted cranioplasty has become a widespread technique to correct scaphocephaly in children with sagittal synostosis, predicting head shape changes induced by the gradual opening of the springs remains challenging. This study aimed to explore the role of cranial bone structure on surgical outcomes. Patients with isolated sagittal synostosis undergoing spring-assisted cranioplasty at GOSH (London, UK) were recruited (n=18, age: 3-8 months). Surgical outcome was assessed by the change in cephalic index measured on 3D head scans acquired before spring insertion and after their removal using a 3D handheld scanner. Parietal bone samples routinely discarded during spring-assisted cranioplasty were collected and scanned using micro-computed tomography. From visual assessment of such scans, bone structure was classified into one- or three-layered, the latter indicating the existence of a diploë cavity. Bone average thickness, volume fraction and surface density were computed and correlated with changes in cephalic index. Cephalic index increased for all patients (p<0.001), but individual improvement varied. Although the patient age and treatment duration were not significantly correlated with changes in cephalic index, bone structural parameters were. The increase of cephalic index was smaller with increasing bone thickness (Pearson's r=-0.79, p<0.001) and decreasing bone surface density (r=0.77, p<0.001), associated with the three-layered bone structure. Variation in parietal bone micro-structure was associated with the magnitude of head shape changes induced by spring-assisted cranioplasty. This suggests that bone structure analysis could be a valuable adjunct in designing surgical strategies that yield optimal patient-specific outcomes.

Improving fit of bicycle helmet liners using 3D anthropometric data

3D anthropometry has provided much-needed information about the size and shape of the head, which can be used to improve the fit of protective helmets. In this study, a new 3D head scan sizing method was implemented in a reverse engineering approach for bicycle helmet liner dimensioning. The inside liner of a commercially available helmet was modified to improve the fit for a selected size group of 30 participants. The fit of the standard and new liner were assessed and compared, using the Helmet Fit Index (HFI). The HFI scores showed a significant improvement of overall fit (Difference: 11.32 ± 7.82 (μ ± SD), p < 0.0005) and for each of five defined regions of the liner inside surface. The presented methodology for dimensioning helmet liners based on 3D anthropometry proved effective, resulting in improved fit for the end users.

Application of massive 3D head and facial scan datasets in ergonomic head-product design

3D human body scan datasets have been collected and various techniques of post-processing, size and shape analysis, and applications in product design have been introduced. This paper is aimed to introduce techniques and cases of head-product design studies based on massive 3D head scan datasets. 3D scan datasets of the head, face and ear were collected and post-processed by editing, landmarking, alignment and measurement. For useful application of the 3D head scan datasets in head-product designs, we applied several analysis techniques such as development of a sizing system, selection of representative models, analysis of shape variation based on template-registered 3D scans and analysis of virtual fit. This paper can help understand an overview of techniques regarding massive 3D scan datasets and their applications to ergonomic product design.

Application of massive 3D head and facial scan datasets in ergonomic head-product design

Application of massive 3D head and facial scan datasets in ergonomic head-product design

Variations in Head-and-Face Shape of Chinese Civilian Workers

This study aims to elucidate variations in head-and-face shape among the Chinese civilian workers. Most respirator manufacturers are using outdated, Western anthropometric data to design respirators for the Chinese workers. Therefore, newly acquired anthropometric data specific to the Chinese population are needed to create more effective personal protective equipment. The three-dimensional (3D) head scans of 350 participants, who were selected from the 3000 participants in the 2006 Chinese Anthropometric Survey, were processed using geometric processing techniques. Each scan was then linked with the others, making statistical shape analysis on a dense set of 3D points possible. Furthermore, this provided for the reduction of scan noise as well as for the patching of holes. Following general scan correspondence and fine tuning, principal component analysis was used to analyze the variability in head-and-face shape of the 3D images. More than 90% of the variability among head-and-face shapes was accounted for with 26 principal components. Future study is recommended so the overall usefulness of the point cloud-based approach for the quantification of variations in facial morphology may be determined.

A comparison of different methods for head measurements

Traditional univariate anthropometric studies document substantial human dimensions with a long history. With the development of new technologies such as laser scanning, the dimensions could be directly extracted from 3D scans. The SizeChina anthropometric survey gathered more than 2,000 3D head scans together with traditional head measurements in China. This study analysed the dimension differences between traditional method and 3D method in terms of head circumference, head length and head breadth. It is found that the results from all dimensions of the 3D scans are larger than the traditional measurements. The differences for females and young group are larger than those of males and older group respectively. These have to be considered when comparing direct measurement results with extracted anthropometric measures from scanned data.

Deriving statistical fit contours and shape of an aerosol mask from 3D head scans

This paper deals with mask shape design to ensure a statistical fit for a given population represented by tessellated 3D head scans. The mask rim curve should contact the face without gaps. This work identifies corresponding points on the mask and facial landmarks. The landmarks define a coordinate frame which helps extract rim contours (mask contours) and facial midline contours (profile contours) from the 3D model. Bilateral symmetrisation procedures are used to correct mild asymmetry in the mask contours. The statistical distribution induced by the contours on a grid is used to derive percentile contour geometries. Mask shapes generated by stacking elliptic cross sections, scaling and convex hull have been explored. 5th, 50th and 85th percentile masks are used for visual fit evaluation on a set of representative faces. The procedure establishes that the landmark-based approach provides a systematic method not only for design but also for fit evaluation.

Shape Analysis of 3D Head Scan Data for U.S. Respirator Users

In 2003, the National Institute for Occupational Safety and Health (NIOSH) conducted a head-and-face anthropometric survey of diverse, civilian respirator users. Of the 3,997 subjects measured using traditional anthropometric techniques, surface scans and 26 three-dimensional (3D) landmark locations were collected for 947 subjects. The objective of this study was to report the size and shape variation of the survey participants using the 3D data. Generalized Procrustes Analysis (GPA) was conducted to standardize configurations of landmarks associated with individuals into a common coordinate system. The superimposed coordinates for each individual were used as commensurate variables that describe individual shape and were analyzed using Principal Component Analysis (PCA) to identify population variation. The first four principal components (PC) account for 49% of the total sample variation. The first PC indicates that overall size is an important component of facial variability. The second PC accounts for long and narrow or short and wide faces. Longer narrow orbits versus shorter wider orbits can be described by PC3, and PC4 represents variation in the degree of ortho/prognathism. Geometric Morphometrics provides a detailed and interpretable assessment of morphological variation that may be useful in assessing respirators and devising new test and certification standards.

Rapid preliminary helmet shell design based on three-dimensional anthropometric head data

The ergonomic design of helmets is very important for those who wear them for long periods on the job; for example, construction workers and security personnel. A helmet's weight, stability, and ability to protect are especially important. According to our case study, fitting design according to head shape can significantly reduce the weight and enhance the stability of a helmet. The traditional helmet design process takes a long time, and is thus unsuitable for individually customized shape design. In this paper, a rapid preliminary design method for the helmet shell and a corresponding toolkit are introduced, taking advantage of three-dimensional (3D) anthropometric head scans. A 3D head model is first generated from the 3D head scan of the intended user or representative user of an intended population group. Then a semi-parametric surface modelling tool is applied to quickly generate the helmet shell by simply inputting several parameters related to helmet protection, size, and shape requirements and adjusting several key curves. In a case study, the new design by the proposed method and the existing design by the traditional method were compared with regard to weight, centroid, and moments of inertia to demonstrate the effectiveness of the proposed method.