A hybrid photoacoustic-ultrasound breast imager - From the lab towards the clinic
Due to the COVID-19 crisis the PhD defence of Maura Dantuma will take place (partly) online.
The PhD defence can be followed by a live stream.
Maura Dantuma is a PhD student in the research group Multi-Modality Medical imaging (M3I). Her supervisors are prof.dr. S. Manohar and prof.dr. L.Oei from the Faculty of Science & Technology (S&T).
The current diagnostic pathway for breast cancer is cumbersome as very often two and sometimes three different imaging modalities need to be used to come to a diagnosis. Pathologic assessment of resected tissue specimens is seen as the golden standard and is performed on top of imaging to confirm the diagnosis. A non-invasive imaging modality with a high sensitivity and specificity for all breast lesion types, which can be used as a stand-alone technique and replaces the currently used modalities, can be considered as the holy grail in breast cancer screening and diagnosis. The introduction of such a modality can significantly reduce healthcare costs and also lower the burden to the patient.
Photoacoustic breast imaging has not proven itself to come close to be this modality, but is one of the techniques that is being researched with high potential. This technique can visualize the blood vessels inside the breast, by illuminating the breast with laser light and recording acoustic waves that are generated in response to the absorbed optical energy by the chromophore haemoglobin in the blood. Abnormal vessel architectures or high vessel densities in the photoacoustic image can potentially give away the presence of cancer.
To elevate PA breast imaging to a modality that can have a strong impact in patient care, several issues require attention. This thesis starts with an overview of existing photoacoustic breast imaging systems and the issues that still have to be addressed are discussed. With all this knowledge in mind, a new photoacoustic breast imaging system was designed and developed by the European Horizon 2020 PAMMOTH consortium. This system, called the PAMMOTH system, is the first multi-modal multi-spectral photoacoustic-ultrasound system with a hemispherical detection array for full breast imaging in 3D. With the combination of the photoacoustic and ultrasound modalities we hope to acquire much information about the tumour morphology and physiology to thereby hopefully be successful in breast cancer diagnosis.
The aim of this thesis was to prepare the grounds for a clinical feasibility study with this PAMMOTH system. The system was designed as an all-round system, meaning that parameters in the measurement protocol, which affect the imaging performance and the measurement times, could be programmed in the control software. Systematic tests were be done to find a measurement sequence resulting in the best imaging quality. For this testing phase, a set of test objects and a breast mimicking phantom were designed and developed, to enable testing of individual system modules and optimization of the measurement protocol via assessment of the image quality metrics as a function of settings in the measurement sequence and optical and acoustic inversion models. This thesis describes both the production protocols and applicabilities of the test objects and phantoms as the testing that was performed with them. In addition to the system testing phase, also safety tests were performed and the study protocol for the pre-clinical study was prepared to get approval to start with measurements in the clinic. The first in vivo imaging results acquired as part this clinical study are were also used as part of the system testing and optimization phase and are also presented in this thesis.