Project Description

The project SCOPIA – Development of software supported clinical devices based on endoscope technology focuses on the development of diagnostic methods and clinical devices based on the results of basic and industrial research. Within the frame of the project, besides the development of mechanical devices, new light sources will be developed. The results of the project will be used in clinical fields, such as cardiology (preserving the heart valves) gynaecology (In vitro fertilisation – IVF) and oncology (pleural biopsy).

Aortic valve-sparing surgery

In Hungary cardiovascular diseases are the most common causes of death, and a common disease of the heart valves’ is the aortic valve insufficiency. During the previous decades the only form of therapy was the replacement of the diseased aortic valve with an artificial aortic valve, which unfortunately resulted in plenty of complications. In our days the focus is on a surgical intervention which spares the aortic valve, thus the development of complications can be avoided. During valve-sparing surgery surgeons presume that the three cusps or leaflets of the aortic valves make equal angles, although their previous anatomical studies contradict this. As in reality aortic valves do not make equal angles, we decided to develop a new measuring device to be applied during the operation. This device will enable the surgeon to measure the angles between the cusps and to reproduce the angles during the aortic valve-sparing surgery. As the surgery requires high precision, several tasks related to geometrical modelling, surgery planning and to the simulation of the blood flow through the heart valves need to be solved to produce a prototype. Thus, the surgical procedure can be validated.

 

Enhancing the chances of the success of the Embryo transfer

As it is demonstrated in the corresponding literature, doctors can receive more information about human tissue by using multispectral light sources beside the usual white light source. By making use of this observation, one of the aims of the project is to measure the general condition of the endometrium, so that the most appropriate time for the successful Embryo transfer can be determined. To achieve this, we plan to produce a multispectral light source, which supports image diagnostics, and to integrate it into a generally used hysteroscope. According to our expectations, if the necessary light force is ensured, this multispectral light source will be able to produce the spectrum of near-infrared light and visible light. With the help of the light source, the most relevant wave-lengths can be determined by applying machine learning methods. The necessary image describers will be obtained from the learning videos made by the whole spectrum. In order to achieve our goals we will have to solve several problems concerning video improvement. A further challenge is the efficient processing of Big Data and the real-time decision support. We plan to validate the developed hysteroscope in clinical environment.

Multimodal virtual demonstration software used at bronchoscopy

One of the important steps in determining the stage of malignant lungs tumours is the precise evaluation of the status of the lymph nodes (nodi lymphatici). During the course of a usual diagnostic procedure images are made of the lungs through contrast enhanced Computed Tomography (CT) and, if it is required, then images are taken through Fluorodeoxyglucose Positron Emission Tomography (FDG-PET). However, defining the standard values is not enough to characterize the tumorous tissues, so tissue samples need to be taken via endoscopy through the windpipe (trachea). The hit probability in the case of interventions made on the basis of evidences produced in a traditional way is only 60%. So that the intervention can be made more efficient, several methods have been developed which fail to spread because of the high expenses of their application. Recently, the combination of PET/CT and CT-based virtual bronchoscopy have been tried out. Within the frame of the project we aim to develop multimodal virtual demonstration software used at bronchoscopy which enables doctors to plan the surgical biopsies through automated processing of CT and PET/CT imaged. Simultaneously with the development of the software, we will solve the problem of the real-time registering of the segmentation of respiratory system based on anatomical model, the virtual space and the video records made during the surgical intervention. Furthermore, by the registration of the modalities, we will develop a navigation system for tissue sampling from the expected location of the tumour.