ABSTRACT
OBJECTIVE: To generate physical models of fetuses from images obtained with three-dimensional ultrasound (3D-US), magnetic resonance imaging (MRI), and, occasionally, computed tomography (CT), in order to guide additive manufacturing technology. MATERIALS AND METHODS: We used 3D-US images of 31 pregnant women, including 5 who were carrying twins. If abnormalities were detected by 3D-US, both MRI and in some cases CT scans were then immediately performed. The images were then exported to a workstation in DICOM format. A single observer performed slice-by-slice manual segmentation using a digital high resolution screen. Virtual 3D models were obtained from software that converts medical images into numerical models. Those models were then generated in physical form through the use of additive manufacturing techniques. RESULTS: Physical models based upon 3D-US, MRI, and CT images were successfully generated. The postnatal appearance of either the aborted fetus or the neonate closely resembled the physical models, particularly in cases of malformations. CONCLUSION: The combined use of 3D-US, MRI, and CT could help improve our understanding of fetal anatomy. These three screening modalities can be used for educational purposes and as tools to enable parents to visualize their unborn baby. The images can be segmented and then applied, separately or jointly, in order to construct virtual and physical 3D models.
OBJETIVO: Gerar modelos físicos de fetos utilizando imagens obtidas por ultrassonografia tridimensional (US3D), ressonância magnética (RM) e, em alguns casos, tomografia computadorizada (TC), para orientar a técnica de adição de camadas. MATERIAIS E MÉTODOS: Foram usadas imagens obtidas de 31 gestantes, incluindo 5 casos de gestação gemelar. Os exames foram realizados usando US3D, RM e em alguns casos TC, e os arquivos foram exportados para uma estação de trabalho em formato DICOM. Um único observador realizou o processo de segmentação manual usando tela de alta resolução. Um software que converte imagens médicas em modelos numéricos foi utilizado para construir modelos virtuais 3D, que foram fisicamente materializados. RESULTADOS: Os modelos virtuais e físicos baseados na US3D, RM e TC realizados separadamente ou em conjunto foram concluídos com sucesso. A aparência pós-natal do feto abortado ou do recém-nascido se assemelhou muito com os modelos físicos, particularmente nos casos de malformações. CONCLUSÃO: O uso da US3D, RM e TC pode ajudar para melhor compreensão das características físicas do feto. Essas técnicas podem ser usadas com fins didáticos para auxiliar na abordagem multidisciplinar e na melhor compreensão dos pais. As imagens podem ser segmentadas e aplicadas separadamente ou combinadas para construir modelos virtuais 3D e físicos.
ABSTRACT
BACKGROUND: Twin-to-twin transfusion syndrome (TTTS) is a serious complication of monochorionic twin gestation, which has a high risk of perinatal morbidity and mortality. Fetoscopic laser photocoagulation of the vascular anastomoses is the preferred treatment. Nowadays, Solomon technique which consists the vascular coagulation of all anastomoses in the placental equator is the preferable method. OBJECTIVE: To develop a method of virtual fetoscopy by means of ultrasound and resonance magnetic imaging (MRI) scan data to allow the identification of placental equator. RESULTS: MRI was performed in a twin monochorionic twin pregnancy with 28 weeks previously treated for TTTS with fetoscopic laser photocoagulation. A three-dimensional (3D) file of the placenta was created from overlapping image layers generated by MRI using Mimics software. Virtual fetoscopy allowed adequate visualization of fetuses, umbilical cord insertion and placental equator. CONCLUSION: Virtual fetoscopy is a non-invasive technique which allowed adequate identification of placental equator, and it may be an important learning method to novice fetal endoscopic surgeons in TTTS cases.