Search

Breast cancer is the leading cause of cancer death among women. Accordingly, appropriate screening protocols for early diagnosis and treatment are crucial to fight against this disease. Artificial intelligence (AI) plays a significant role in screening mammograms and helping radiologists identify potential abnormalities. When a breast anomaly is detected, a biopsy is taken and a tissue marker is placed in the affected area, serving as a reference point for future treatments. To ensure reliability, it is crucial to comprehend how these tissue markers contribute to the process of breast cancer detection using AI systems. To this end, this study initially presents an image processing algorithm that facilitates the creation of an annotated dataset comprising images with tissue markers. This dataset is then used to develop a deep learning approach that is capable of discriminating mammograms with tissue markers, as well as of segmenting these objects. The study concludes with a methodology for analyzing the performance of an AI system in breast cancer screening. This framework does a comparison between the AI system's detected anomalous regions and the locations of tissue markers. The results highlight the strong performance of the deep learning model in both the segmentation and detection of mammograms with tissue markers. Moreover, the findings demonstrated that regardless of the presence of clips, AI systems can identify potential abnormalities with a reduced probability of marking tissue markers.

Cerebral small vessel disease causes cognitive impairment, dementia, and stroke and is characterized by white matter hyperintensities (WMH) and other brain lesions. Lesion-symptom mapping (LSM) aims to understand the relationship between brain lesion location and cognition by identifying strategic lesion locations. This study presents a multi-output deep learning lesion-symptom mapping (DL-LSM) approach using explainable artificial intelligence (XAI). This approach is validated in a simulation study using WMH segmentations of 821 memory clinic patients and artificial cognitive scores. The study comprised three experiments. The first involved generating artificial cognitive scores based on the lesion load within three predefined regions of interest (ROIs). The second experiment studied the impact of adding noise to these scores on the DL and XAI methods. The third experiment explored whether intercorrelations between different ROIs in the artificial cognitive scores could be detected. Two convolutional neural networks (CNN) were developed to predict the artificial cognitive scores, and XAI was used to identify the locations that influenced these predictions. The methods were evaluated by quantifying the model’s predictive performance, identifying the ROIs in the XAI's attribution maps, and quantifying the intercorrelation of the detected ROIs. This study demonstrates that DL models can predict multiple artificial cognitive scores based on WMH segmentations, and that XAI can identify the ROIs associated with the simulated cognitive scores. Additionally, the results demonstrate that DL-LSM is robust to low levels of noise in the artificial cognitive scores and can detect intercorrelations between ROIs. These findings indicate that DL and XAI can be used to perform LSM in order to predict cognitive scores and determine their relationship with specific lesion locations.