The TeleUSG project is carried out in collaboration with the University Hospital of Geneva HUG and aims to improve a health telemonitoring services provided to healthcare practitioners, trained nurses, as well as their patients, in developing nations of African and South American countries. In this project we use an integrated medical hardware and software tool co-called Voyager, provided by Ardent Sound, Inc. Voyager records a sequence of ultrasonography (i.e., USG) images. In the existing service deployment, healthcare practitioners and trained nurses from remote areas are able to examine the patient and pre-record ultrasonography images from his/her and then upload these images to a central storage, from where they could be accessed by a specialist - radiologists (e.g., in Geneva), aiming to (tele-) diagnose the patient.
However, this process is completely asynchronous, which hinders the quality of the tele-diagnosis service as follows: a) the diagnosis can be too late, with respect to the criticality of the patient’s health state; b) the quality of the recorded images can be insufficient for a reliable diagnosis, leading to a misdiagnosis, or a need for a repeated examination of the patient; c) the remote practitioner or a nurse performing the examination do not get a direct feedback on their examination skills, hence may find it difficult to learn how to perform an effective examination enabling to a reliable diagnosis.
To overcome the current approach, the goal of the TeleUSG project is to improve the efficiency and effectiveness of the examination and facilitate the reliability of diagnosis, by using the same hardware and expanding the software component towards a real-time, continuous exchange of ultrasonography images between the remote practitioners performing the examination and the specialists. Additionally, we aim to establish a real-time, continuous voice communication service between these service users, enabling the remote practitioner or a nurse performing the examination to be provided with a real-time feedback from the radiologist with respect to the examination and zones of interest, for which more detailed ultrasonography images need to be gathered.
Furthermore, the engineering challenge being tackled when following this approach raises from the fact that the patient’s examination takes place in many cases in poorly connected, bandwidth-constrained and jittery environments (e.g., including satellite-based access links). Therefore, our goal is to develop an adaptive application protocol that would always “fit” the volume of data for ultrasonography images and voice communication, to the currently available bandwidth along the end-to-end communication path between the patient’s examination and the radiologist.