Confronting historical legacies of biological anthropology in South Africa-Restitution, redress and community-centered science: The Sutherland Nine.

We describe a process of restitution of nine unethically acquired human skeletons to their families, together with attempts at redress. Between 1925-1927 C.E., the skeletonised remains of nine San or Khoekhoe people, eight of them known-in-life, were removed from their graves on the farm Kruisrivier, near Sutherland in the Northern Cape Province of South Africa. They were donated to the Anatomy Department at the University of Cape Town. This was done without the knowledge or permission of their families. The donor was a medical student who removed the remains from the labourers' cemetery on his family farm. Nearly 100 years later, the remains are being returned to their community, accompanied by a range of community-driven interdisciplinary historical, archaeological and analytical (osteobiographic, craniofacial, ancient DNA, stable isotope) studies to document, as far as possible, their lives and deaths. The restitution process began by contacting families living in the same area with the same surnames as the deceased. The restitution and redress process prioritises the descendant families' memories, wishes and desire to understand the situation, and learn more about their ancestors. The descendant families have described the process as helping them to reconnect with their ancestors. A richer appreciation of their ancestors' lives, gained in part from scientific analyses, culminating with reburial, is hoped to aid the descendant families and wider community in [re-]connecting with their heritage and culture, and contribute to restorative justice, reconciliation and healing while confronting a traumatic historical moment. While these nine individuals were exhumed as specimens, they will be reburied as people.

Reconstruction of the mandible from partial inputs for virtual surgery planning.

Statistical Shape Models (SSMs) and Sparse Prediction Models (SPMs) based on regressions between cephalometric measurements were compared against standard practice in virtual surgery planning for reconstruction of mandibular defects. Emphasis was placed on the ability of the models to reproduce clinically relevant metrics. CT scans of 50 men and 50 women were collected and split into training and testing datasets according to an 80:20 ratio. The scans were segmented, and anatomical landmarks were identified. SPMs were constructed based on direct regressions between measurements derived from the anatomical landmarks. SSMs were developed by establishing correspondence between the segmented meshes, performing alignment, and principal component analysis. Anterior and bilateral defects were simulated by removing sections of the mandibles in the testing set. Measurement errors after reconstruction ranged from 1.07˚ to 2.2˚ and 0.66 mm to 2.02 mm for mirroring, from 0.45˚ to 3.67˚ and 0.66 mm to 2.54 mm for the SSMs, and from 1.74˚ to 5.01˚ and 0.64 mm to 2.89 mm for the SPMs. Surface-to-surface errors ranged from 1.01 mm to 1.29 mm and 1.06 mm to 1.33 mm for mirroring and SSMs, respectively. Based on the results, SSMs are recommended for VSP in the absence of normal patient anatomy.

Dynamic multi feature-class Gaussian process models.

In model-based medical image analysis, three relevant features are the shape of structures of interest, their relative pose, and image intensity profiles representative of some physical properties. Often, these features are modelled separately through statistical models by decomposing the object's features into a set of basis functions through principal geodesic analysis or principal component analysis. However, analysing articulated objects in an image using independent single object models may lead to large uncertainties and impingement, especially around organ boundaries. Questions that come to mind are the feasibility of building a unique model that combines all three features of interest in the same statistical space, and what advantages can be gained for image analysis. This study presents a statistical modelling method for automatic analysis of shape, pose and intensity features in medical images which we call the Dynamic multi feature-class Gaussian process models (DMFC-GPM). The DMFC-GPM is a Gaussian process (GP)-based model with a shared latent space that encodes linear and non-linear variations. Our method is defined in a continuous domain with a principled way to represent shape, pose and intensity feature-classes in a linear space, based on deformation fields. A deformation field-based metric is adapted in the method for modelling shape and intensity variation as well as for comparing rigid transformations (pose). Moreover, DMFC-GPMs inherit properties intrinsic to GPs including marginalisation and regression. Furthermore, they allow for adding additional pose variability on top of those obtained from the image acquisition process; what we term as permutation modelling. For image analysis tasks using DMFC-GPMs, we adapt Metropolis-Hastings algorithms making the prediction of features fully probabilistic. We validate the method using controlled synthetic data and we perform experiments on bone structures from CT images of the shoulder to illustrate the efficacy of the model for pose and shape prediction. The model performance results suggest that this new modelling paradigm is robust, accurate, accessible, and has potential applications in a multitude of scenarios including the management of musculoskeletal disorders, clinical decision making and image processing.

Cross-Modality Image Adaptation Based on Volumetric Intensity Gaussian Process Models (VIGPM).

Image-based diagnosis routinely depends on more that one image modality for exploiting the complementary information they provide. However, it is not always possible to obtain images from a secondary modality for several reasons such as cost, degree of invasiveness and non-availability of scanners. Three-dimensional (3D) morphable models have made a significant contribution to the field of medical imaging for feature-based analysis. Here we extend their use to encode 3D volumetric imaging modalities. Specifically, we build a Gaussian Process (GP) over transformations establishing anatomical correspondence between training images within a modality. Given, two different modalities, the GP's eigenspace (latent space) can then be used to provide a parametric representation of each image modality, and we provide an operator for cross-domain translation between the two. We show that the latent space yields samples that are representative of the encoded modality. We also demonstrate that a 3D volumetric image can be efficiently encoded in latent space and transferred to synthesize the corresponding image in another modality. The framework called VIGPM can be extended by designing a fitting process to learn an observation in a given modality and performing cross-modality synthesis. Clinical Relevance- The proposed method provides a way to access a multi modality image from one modality. Both the source and synthetic modalities are in anatomical correspondence giving access to registered complementary information.

Assessment of the implementation context in preparation for a clinical study of machine-learning algorithms to automate the classification of digital cervical images for cervical cancer screening in resource-constrained settings.

Introduction: We assessed the implementation context and image quality in preparation for a clinical study evaluating the effectiveness of automated visual assessment devices within cervical cancer screening of women living without and with HIV. Methods: We developed a semi-structured questionnaire based on three Consolidated Framework for Implementation Research (CFIR) domains; intervention characteristics, inner setting, and process, in Cape Town, South Africa. Between December 1, 2020, and August 6, 2021, we evaluated two devices: MobileODT handheld colposcope; and a commercially-available cell phone (Samsung A21ST). Colposcopists visually inspected cervical images for technical adequacy. Descriptive analyses were tabulated for quantitative variables, and narrative responses were summarized in the text. Results: Two colposcopists described the devices as easy to operate, without data loss. The clinical workspace and gynecological workflow were modified to incorporate devices and manage images. Providers believed either device would likely perform better than cytology under most circumstances unless the squamocolumnar junction (SCJ) were not visible, in which case cytology was expected to be better. Image quality (N = 75) from the MobileODT device and cell phone was comparable in terms of achieving good focus (81% vs. 84%), obtaining visibility of the squamous columnar junction (88% vs. 97%), avoiding occlusion (79% vs. 87%), and detection of lesion and range of lesion includes the upper limit (63% vs. 53%) but differed in taking photographs free of glare (100% vs. 24%). Conclusion: Novel application of the CFIR early in the conduct of the clinical study, including assessment of image quality, highlight real-world factors about intervention characteristics, inner clinical setting, and workflow process that may affect both the clinical study findings and ultimate pace of translating to clinical practice. The application and augmentation of the CFIR in this study context highlighted adaptations needed for the framework to better measure factors relevant to implementing digital interventions.

Review of 2-D/3-D Reconstruction Using Statistical Shape and Intensity Models and X-Ray Image Synthesis: Toward a Unified Framework.

Patient-specific three-dimensional (3-D) bone models are useful for a number of clinical applications such as surgery planning, postoperative evaluation, as well as implant and prosthesis design. Two-dimensional-to-3-D (2-D/3-D) reconstruction, also known as model-to-modality or atlas-based 2-D/3-D registration, provides a means of obtaining a 3-D model of a patient's bones from their 2-D radiographs when 3-D imaging modalities are not available. The preferred approach for estimating both shape and density information (that would be present in a patient's computed tomography data) for 2-D/3-D reconstruction makes use of digitally reconstructed radiographs and deformable models in an iterative, non-rigid, intensity-based approach. Based on a large number of state-of-the-art 2-D/3-D bone reconstruction methods, a unified mathematical formulation of the problem is proposed in a common conceptual framework, using unambiguous terminology. In addition, shortcomings, recent adaptations, and persisting challenges are discussed along with insights for future research.

Precision assessment of stereo-photogrammetrically derived facial landmarks in infants.

BACKGROUND: In order to examine the suitability of an anthropometric imaging device for a particular study population, it is necessary to assess measurement error on members of the population. AIM: We provide an analysis of the precision of a stereo-photogrammetric tool in locating 3D landmark coordinates on images of infants and present the results in a way that facilitates direct comparison with some of the precision results found in the literature. SUBJECTS AND METHODS: Intra- and inter-observer precision are examined for 26 landmarks on 21 infants. Mean absolute differences for landmark coordinates and linear distances, mean landmark coordinate errors, and relative errors of magnitude are calculated. RESULTS: Relative error of magnitude is less than 4% for the majority of intra- and inter-observer inter-landmark distance comparisons. Landmarks lying on the midline of the face and on smooth regions with gentle curvature, in particular the sellion, glabella and nasion, have relatively low precision. CONCLUSION: Our precision values compare well with those reported in published studies on landmark location in 3D surface images.

A review of facial image analysis for delineation of the facial phenotype associated with fetal alcohol syndrome.

The facial anomalies associated with fetal alcohol syndrome (FAS), some of which are also present in individuals with less severe forms of the broader category of fetal alcohol spectrum disorders (FASD), are typically identified with the aid of linear distance measurements taken between facial landmarks. Digital facial imaging methods are increasingly being used in syndrome delineation. Distance measurements derived from stereo-photogrammetry and facial surface imaging have been used to study the FAS facial anomalies. Geometric morphometric methods capture the spatial arrangement between landmarks, providing a statistical platform for comparison of facial shapes, and have been shown to hold promise for characterizing the FAS facial shape. We review the progression in the use of imaging and image analysis methods in studies on the facial phenotype associated with FAS.

Morphometric analysis and classification of the facial phenotype associated with fetal alcohol syndrome in 5- and 12-year-old children.

Landmark-based morphometric analysis holds promise for quantitative assessment of craniofacial dysmorphology. We describe an application of facial shape analysis to characterize the facial anomalies associated with fetal alcohol syndrome (FAS) in a mixed ancestry population. Generalized Procrustes analysis, regression and discriminant function analysis were applied to stereo-photogrammetrically derived 3D coordinates of landmarks taken from 34 subjects (n = 17 FAS and n = 17 normal controls). Four shape analyses were carried out, namely a comparison of the FAS and control facial shapes at age 5, and one at age 12; a comparison of the FAS facial shapes at ages 5 and 12; and a comparison of control facial shapes at ages 5 and 12. The first two analyses showed that the FAS face is characterized by small palpebral fissures, a thin upper lip, and midfacial hypoplasia. Classification of subjects as having FAS using leave-one-out cross-validation showed that the 5-year-old group could be classified with 95.46% accuracy and the 12-year-olds with 80.13% accuracy. The third and fourth analyses revealed that the differences in facial shape between FAS individuals in different age groups were more pronounced than for control individuals, supporting the notion that FAS facial anomalies diminish with age. Geometric morphometric analysis of stereo-photogrammetrically derived 3D facial landmarks allows visualization of the facial anomalies associated with FAS, as well as classification of facial shapes.

Design, construction, and testing of a stereo-photogrammetric tool for the diagnosis of fetal alcohol syndrome in infants.

Stereo-photogrammetry provides a low cost, easy to use, and noninvasive alternative to traditional facial anthropometry for the diagnosis of fetal alcohol syndrome (FAS). We describe such a system for use in obtaining 3-D facial information in infants. The infant is photographed using three high resolution digital cameras simultaneously while seated in a car seat. The subject's head is enclosed in a control frame during imaging. Technical system tests, namely control frame interpolation, camera calibration reliability, and camera synchronization delay assessments were performed. Direct and stereo-photogrammetric measurements of a doll were compared. Of 275 inter-landmark distances, 100% were within a 1.5 mm error range and 92.36% within a 1 mm error range when the two modalities were compared. Stereo-photogrammetry proved to be highly precise with submillimeter error in landmark placement for all landmarks on the doll. An intra-modality comparison of inter-landmark distances using two sets of images of five subjects showed the stereo-photogrammetric system to be highly reliable, with an average 72.25% of distances within a 1 mm error range. The system has potential for large scale screening and surveillance studies for FAS.