Nanna M. Sijtsema

Alessia De Biase, Nanna Maria Sijtsema, Lisanne van Dijk et al.

Tumor segmentation is a fundamental step for radiotherapy treatment planning. To define an accurate segmentation of the primary tumor (GTVp) of oropharyngeal cancer patients (OPC), simultaneous assessment of different image modalities is needed, and each image volume is explored slice-by-slice from different orientations. Moreover, the manual fixed boundary of segmentation neglects the spatial uncertainty known to occur in tumor delineation. This study proposes a novel automatic deep learning (DL) model to assist radiation oncologists in a slice-by-slice adaptive GTVp segmentation on registered FDG PET/CT images. We included 138 OPC patients treated with (chemo)radiation in our institute. Our DL framework exploits both inter and intra-slice context. Sequences of 3 consecutive 2D slices of concatenated FDG PET/CT images and GTVp contours were used as input. A 3-fold cross validation was performed three times, training on sequences extracted from the Axial (A), Sagittal (S), and Coronal (C) plane of 113 patients. Since consecutive sequences in a volume contain overlapping slices, each slice resulted in three outcome predictions that were averaged. In the A, S, and C planes, the output shows areas with different probabilities of predicting the tumor. The performance of the models was assessed on 25 patients at different probability thresholds using the mean Dice Score Coefficient (DSC). Predictions were the closest to the ground truth at a probability threshold of 0.9 (DSC of 0.70 in the A, 0.77 in the S, and 0.80 in the C plane). The promising results of the proposed DL model show that the probability maps on registered FDG PET/CT images could guide radiation oncologists in a slice-by-slice adaptive GTVp segmentation.

Erfan Darzi, Florian Dubost, N. M. Sijtsema et al.

Federated learning offers a privacy-preserving framework for medical image analysis but exposes the system to adversarial attacks. This paper aims to evaluate the vulnerabilities of federated learning networks in medical image analysis against such attacks. Employing domain-specific MRI tumor and pathology imaging datasets, we assess the effectiveness of known threat scenarios in a federated learning environment. Our tests reveal that domain-specific configurations can increase the attacker's success rate significantly. The findings emphasize the urgent need for effective defense mechanisms and suggest a critical re-evaluation of current security protocols in federated medical image analysis systems.

Erfan Darzi, Nanna M. Sijtsema, P. M. A van Ooijen

This paper explores the security aspects of federated learning applications in medical image analysis. Current robustness-oriented methods like adversarial training, secure aggregation, and homomorphic encryption often risk privacy compromises. The central aim is to defend the network against potential privacy breaches while maintaining model robustness against adversarial manipulations. We show that incorporating distributed noise, grounded in the privacy guarantees in federated settings, enables the development of a adversarially robust model that also meets federated privacy standards. We conducted comprehensive evaluations across diverse attack scenarios, parameters, and use cases in cancer imaging, concentrating on pathology, meningioma, and glioma. The results reveal that the incorporation of distributed noise allows for the attainment of security levels comparable to those of conventional adversarial training while requiring fewer retraining samples to establish a robust model.

Erfan Darzidehkalani, Nanna M. Sijtsema, P. M. A van Ooijen

Deep learning is effective in diagnosing COVID-19 and requires a large amount of data to be effectively trained. Due to data and privacy regulations, hospitals generally have no access to data from other hospitals. Federated learning (FL) has been used to solve this problem, where it utilizes a distributed setting to train models in hospitals in a privacy-preserving manner. Deploying FL is not always feasible as it requires high computation and network communication resources. This paper evaluates five FL algorithms' performance and resource efficiency for Covid-19 detection. A decentralized setting with CNN networks is set up, and the performance of FL algorithms is compared with a centralized environment. We examined the algorithms with varying numbers of participants, federated rounds, and selection algorithms. Our results show that cyclic weight transfer can have better overall performance, and results are better with fewer participating hospitals. Our results demonstrate good performance for detecting COVID-19 patients and might be useful in deploying FL algorithms for covid-19 detection and medical image analysis in general.

Erfan Darzi, Florian Dubost, Nanna. M. Sijtsema et al.

In this paper, we delve into the susceptibility of federated medical image analysis systems to adversarial attacks. Our analysis uncovers a novel exploitation avenue: using gradient information from prior global model updates, adversaries can enhance the efficiency and transferability of their attacks. Specifically, we demonstrate that single-step attacks (e.g. FGSM), when aptly initialized, can outperform the efficiency of their iterative counterparts but with reduced computational demand. Our findings underscore the need to revisit our understanding of AI security in federated healthcare settings.

Erfan Darzi, Yiqing Shen, Yangming Ou et al.

Optimization-based regularization methods have been effective in addressing the challenges posed by data heterogeneity in medical federated learning, particularly in improving the performance of underrepresented clients. However, these methods often lead to lower overall model accuracy and slower convergence rates. In this paper, we demonstrate that using Vision Transformers can substantially improve the performance of underrepresented clients without a significant trade-off in overall accuracy. This improvement is attributed to the Vision transformer's ability to capture long-range dependencies within the input data.