Anatomical digital twins for medical education: a stepwise guide to create perpetual multimodal three-dimensional reconstruction of digital brain specimens.

Background: The mastery of neuroanatomy is key to medical education, radiological interpretation of neurological signs and symptoms, and ultimately, surgical planning. With the development of imaging technology, the three-dimensional (3D) presentation of anatomical structures has become possible. The plastic models and anatomopathological specimens available for teaching anatomy in medical schools are often obsolete or poorly preserved, and in general, they provide limited margins for an enhanced learning experience, allowing for 3D visualization of the relationship with surrounding structures. To maximize the impact on anatomical teaching, we created a 3D digital model of human brain specimens using computed tomography (CT) and magnetic resonance imaging (MRI) scans, and combined this with the powerful editing capabilities of the open-source 3D Slicer platform for image reconstruction and optimization. Methods: Using cranial specimens donated to scientific research, we first connected the blood vessels and pretreated the specimens with a slow and continuous fluid injection. CT and MRI scans were performed after the injection of the appropriate amount of corresponding contrast agents into the specimens to obtain Digital Imaging and Communications in Medicine (DICOM) images. Subsequently, open-source 3D Slicer software was used to reconstruct the images in three dimensions and edit and optimize them to complete the digital reconstruction of specimens (digital twins). Results: By combining reconstruction modeling of digitized human brain specimens, the intracranial vasculature and the parenchymal anatomy can be largely restored, isolated, and reconstructed through the fusion of multimodal images on the 3D Slicer platform. Since vascular perfusion is better visualized under the CT modality, yet soft tissues such as brain parenchyma are better visualized under the MRI modality, our combined approach provides high-quality 3D model reconstruction. Conclusions: We provide a road map to create a simple digital reconstruction model of human brain specimens. After injection of contrast agent into the specimen vessel, DICOM images are obtained after CT or MRI scanning to visualize vascular reconstruction. After multimodal image data are generated, 3D Slicer software can be used for 3D reconstruction and optimization of the acquired images, thus providing a digital 3D reconstruction model of the cranial brain specimen. This technology can provide observers with more vivid and intuitive 3D images and has a wide range of prospective applications, including the digital preservation of specimen information, medical anatomy teaching, and surgical training.