Hochschule Karlsruhe Hochschule Karlsruhe - University of Applied Sciences
Hochschule Karlsruhe Hochschule Karlsruhe - University of Applied Sciences

Intelligent Systems Research Group

InferMod3D

Development of an Inference Module for Model-Based 3D Reconstruction of People and Body Parts on Portable Devices

Team:
Christian Wernet, M.Sc.
Prof. Dr. Astrid Laubenheimer

Funding: MWK
Duration: 2020 - 2022

Overview

In InferMod3D a medical app for diagnosis and monitoring of plagiocephaly (head asymmetries in infancy) is provided, which should be usable in the pediatric practice as well as by therapists and parents.

AI methods will create 3D reconstructions of infant heads from point clouds. From those reconstructions, measurements for the amount of head deformation such as the Cranial Vault Asymmetry Index (CVAI) can be extractet.

Our method relies on our articulated human shape model JOMS. The research activities within InferMod3D focus on a new approach for the inference, which will be optimized for models without skeletons and should be able to run on mobile devices in real-time.

Results

A prototype of a 3D infant head reconstruction app for mobile phones was developed to act as a cranial asymmetry measurement method.

It uses the iPhone TrueDepth 3D sensor for 3D reconstruction and measurement of an infant's head. This process is challenging due to the uncooperative behaviour of the subject, hence the app provides audible, haptic and visual feedback to support the user during the scanning process. The app is designed to be easy to use following an intuitive user experience design and tested by a team of physical therapists to evaluate the feedback concept and the user experience.

Externer Inhalt

Um diesen Inhalt zu verwenden (Quelle: www.xyz.de), klicken Sie bitte auf Akzeptieren. Wir möchten Sie darauf hinweisen, dass durch die Annahme dieser IFrames Daten an Dritte übertragen oder Cookies gespeichert werden könnten.

Weitere Informationen finden Sie in unserer Datenschutzerklärung.

Related Publications

Rapp F., Meder C., Stoevesandt K., Zeitvogel S., Wernet C., Laubenheimer A.
Feedback mechanisms of an iOS App to record RGBD Data for AI-based Diagnosis and Monitoring of Infant Head Deformation
13th 3DBODY.TECH Conference & Expo, 25-26 Oct. 2022, Lugano, Switzerland. [pdf]

Zeitvogel S., Wernet C., Wetzel J., Laubenheimer A., Stoevesandt K.
RGBD Infant Head Reconstruction for Cranial Vault Asymmetry Estimation
IEEE Access. doi: 10.1109/ACCESS.2022.3160749, March 21, 2022. [pdf]

Partners & Funding

The partners inovex and VARILAG will finally implement the research results in an app and evaluate the quality of the extracted measurements in practice. The medical supply store Storch und Beller kindly provided data for the model building process.

The project is funded by the Ministry of Science, Research and Arts of Baden-Württemberg (MWK).