The invention relates to a process for identification of an implant, optionally worn by an individual, comprising the following steps, performed by a processing unit:

obtaining a radiographic input image in which a dental implant is visible;

classification of a region of interest of said radiographic image by means of at least one convolutional neural network; said classification producing a list of candidate implants to be the implant visible on the image, the list being ordered as a function of a probability of being the implant visible on the image.

A patient's healthcare experience may be enhanced utilizing a system that automatically recognizes the patient based on one or more images of the patient and generates personalized medical assistance information for the patient based on electronic medical records stored for the patient. Such electronic medical records may comprise imagery data and/or non-imagery associated with a medical procedure performed or to be performed for the patient. As such, the imagery and/or non-imagery data may be incorporated into the personalized medical assistance information to provide positioning and/or other types of diagnostic or treatment guidance to the patient or a service provider.

A method including: obtaining three-dimensional (3D) point group by using a measurement result of a 3D sensor; evaluating an influence of noise on the measurement result by using a result obtained by applying a cylinder model expressing each part of a human body to the 3D point group for each part; repeatedly executing a process in which a point group in a cylinder model periphery corresponding to a part in which the influence of noise is determined to be equal to or higher than a threshold is excluded from the 3D point group and the cylinder model is applied again to the 3D point group from which the point group is excluded; and generating a skeleton recognition result by using a result obtained by applying the cylinder model to the 3D point group of a case where the influence of noise on each part is lower than the threshold.

Methods, hardware, and software transform 2D anatomical X-ray images into 3D renderings for surgical preparation. X-ray images of a body part are identified by camera model. A contour is extracted from the X-ray. Each anatomical region of the contour is assigned 2D anatomical values. A separate 3D template for the body part is modified to match the X-ray image by extracting silhouette vertices from the template and their projections. The template is aligned with the x-ray image and projected on an image plane to obtain a 2D projection model. The template is modified to match the anatomical values by comparing the projection with the corresponding anatomical values. Best matching points on the contour for extracted silhouette vertex projections are identified and used to back-project corresponding silhouette vertices. The 3D template is deformed so that its silhouette vertices match the target positions, resulting in a 3D reconstruction for the X-ray image.

A shade selection program is disclosed that predicts the shade choice with the smallest CIEDE2000 color difference for dental composite resin restorations when given a backing and target shade. By utilizing generated regression models, a database of spectral reflectance information, and principles of Kubelka-Munk layering, a highly accurate shade selection program was designed. Additionally, a blending model for quantification of color adjustment potential was developed. Systems and methods for correlating RGB data from the VITA Linearguide 3D Master and VITA Bleached Guide 3D Master shade guides with their spectroradiometric correlates through a regression model while indicating a methodology for validation of accuracy of digital imaging systems are disclosed.

A reference standard or reference system for dental feature location and orientation can include evaluating the facial appearance of one or more subjects based on aesthetic criteria and selecting a subset of subjects. The reference standard further includes constructing 3D representations, (e.g., a 3D virtual model), of the facial and dental features of the subset of subjects, from photographs of the face and mouth of each subject and determining the location and/or orientation of one or more dental features for each subject. The location and/or orientation values for each subject can be used to produce an average value and a standard deviation that forms the basis for a reference standard as part of reference system for evaluating patients. The method and system further includes constructing 3D representations (e.g., a 3D virtual model) of the facial and dental features of a patient from photographs of the face and mouth of the patient and determining the location and/or orientation of one or more dental features of the patient in order to compare them to the reference standard and develop a treatment plan for the subject based on differences between the patient's measurements and reference standard. The reference standard can use the pupils (e.g., in the natural head position orientation) as a landmark for registration and scaling of the reference standard to the patient under evaluation.

A method, apparatus, and system for reliably identifying a specific part of a spine in an image of a human or animal body, includes the steps of determining one or more parts of the spine in the image, determining one or more discriminative parameters for each of the one or more parts of the spine in the image, the discriminative parameters relating to at least one anatomical property of each of the one or more parts of the spine, classifying the discriminative parameters of the one or more parts of the spine in the image, and identifying a specific part of the spine based on the classification of the discriminative parameters of the one or more parts of the spine in the image. An identified vertebra, in particular the T12 vertebra and/or its associated intervertebral discs, can be used advantageously as a starting point of powerful automatic spine labeling algorithms.

The outline of a bone or at least a portion thereof may be determined based on deep neural net and optionally on an active shape model approach. An algorithm may detect a fracture of the bone. An algorithm may also classify the bone fracture and provide guidance on how to treat the fracture.

A medical image processing method using machine learning according to an embodiment of the present invention includes acquiring an X-ray image of an object, identifying a plurality of anatomical regions by applying a deep learning technique for each bone structure region that constitutes the X-ray image, predicting a bone disease according to bone quality for each of the plurality of anatomical regions, and determining an artificial joint that replaces the anatomical region in which the bone disease is predicted.

Improved techniques for determining an object's 3D orientation. An image is analyzed to identify a 2D object and a first set of key points. The first set defines a first polygon. A 3D virtual object is generated. This 3D virtual object has a second set of key points defining a second polygon representing an orientation of the 3D virtual object. The second polygon is rotated a selected number of times. For each rotation, each rotated polygon is reprojected into 2D space, and a matching score is determined between each reprojected polygon and the first polygon. A specific reprojected polygon is selected whose corresponding matching score is lowest. The orientation of the 3D virtual object is set to an orientation corresponding to the specific reprojected polygon. Based on the orientation of the 3D virtual object, an area of focus of the 2D object is determined.