Introduction Breast cancer is the most frequently diagnosed type of cancer in women in 2020. Treatment and prognosis of breast cancer is highly dependent on early and accurate diagnosis. In recent years, many studies have evaluated the use of artificial intelligence for the detection of breast cancer in mammographic images. Another imaging modality, breast-specific gamma imaging (BSGI), or molecular breast imaging, has not yet been subject to AI algorithms to detect breast cancer. In this paper, we aim to develop and evaluate convolutional neural networks (CNNs) that can detect malignancies in breast-specific gamma imaging, and evaluate the efficacy of different machine learning classifiers to classify breast tumors based on estrogen receptor (ER) status, progesterone receptor (PR) status and human epidermal growth factor receptor 2 (HER2-neu) status. Methods Three CNNs were created and trained and tested on a total of 3,503 BSGI images. The models varied in complexity in terms of convolutional layers and filter sizes. A semi quantitative lesion segmentation was created based on adaptive thresholding and shape and location analysis. Radiomics features were extracted, and univariate feature selection was applied to disregard redundant features. Different machine learning classifiers, which are widely used in literature for binary classification problems, were evaluated. Results In detecting malignancies in a dataset containing clean breasts and breast with malignant lesions, the best performing network reached an area under the receiving operating characteristic (AUROC) of 0.93, while an AUROC of 0.88 was achieved when using the same networks in the classification of malignant versus benign lesions. The best performing machine learning classifiers were the linear discriminating analysis (LDA) classifier for ER and PR status, reaching accuracies of 75% in both receptors. In Her2-neu prediction using machine learning, the best accuracy of 69% was achieved by the RF classifier. Discussion & conclusions Based on the results presented in this paper, CNNs can accurately detect malignancies in BSGI images, and discriminate malignancies from benign lesions to a certain extent. The combination of radiomics and machine learning, however, is in its current implementation not accurate enough to predict the ER, PR and Her2-neu status in BSGI images. Future research, however, could focus on using a combination of imaging modalities, such as BSGI, MRI and mammography to improve the predictive accuracy of machine learning.

The aim of this project was the implementation of optical coherence tomography (OCT), an invasive intracoronary imaging (ICI) tool used for the visualization of coronary arteries. The implementation was performed in a structured manner by: (1) A literature review into the current applications of OCT in relation to current standards; (2) A thesis feasibility study into clinical practices, professional views and other processes regarding ICI; (3) A Master’s thesis including a cost-effectiveness analysis (CEA) and the actual implementation of OCT at the department. The project started with a literature review about the current applications of OCT in relation to intravascular ultrasound (IVUS) and coronary angiography (CAG). The literature review has been published in Cardiology and Therapy. In general, OCT-guidance seems to contribute to favorable clinical outcomes compared with CAG-guidance alone. However, OCT-guidance results in similar clinical outcomes as IVUS-guidance. OCT could be considered for lumen assessment and stent related morphology in more complex cases in which CAG interpretation remains uncertain. Since OCT and IVUS have distinct characteristics, these techniques are complementary and should be considered carefully for each patient case based on the benefits and limitations of both techniques. The thesis feasibility study aimed at acquiring insights into the clinical practices, the different views, and other processes (data, training, etcetera) related to ICI at the LUMC and other medical centers. This was achieved by observing clinical practices at the LUMC and other medical centers, and by interviewing clinicians and technicians at those centers. Both the literature review and the feasibility study provided extensive insights in current practices and controversies regarding ICI. Such insights were essential before implementation of OCT. These insights were used to carefully execute the implementation, while accounting for the interests of all stakeholders as much as possible. In addition, the knowledge gained from the literature review and feasibility study were used as input for the master’s thesis. The master’s thesis included a health economic evaluation comparing cost-effectiveness of OCT with IVUS and CAG, and the actual implementation of OCT at the department. The CEA was performed using the concept of Markov modelling. In general, OCT along with CAG was a more cost-effective intracoronary treatment strategy than IVUS along with CAG for patients with CAD. However, the gain in QALYs and reduction in costs were limited. Both IVUS and OCT outperform CAG alone in terms of cost-effectiveness. With this thesis the current applications of OCT in relation to IVUS and CAG were identified, insights into different views and practices regarding ICI in the interventional cardiology domain were obtained and the cost-effectiveness of OCT-guided PCI compared with IVUS-guided PCI and PCI solely guided by CAG was assessed. In addition, OCT was implemented at the department of interventional cardiology of the LUMC. The processes, procedures and considerations around the implementation of OCT at the department of interventional cardiology were identified and described. Finally, the thesis was concluded with an argument about the added value of the technical physician as a new healthcare professional.