arXiv subjects

David Horowitz

Publications and source records attributed to David Horowitz.

3 recordsLinked to original sources

arXiv2026

Foundation Model-guided Iteratively Prompting and Pseudo-Labeling for Partially Labeled Medical Image Segmentation

Automated medical image segmentation has achieved remarkable progress with fully labeled data. However, site-specific clinical priorities and the high cost of manual annotation often yield scans with only a subset of organs labeled, leading to the partially labeled problem that degrades performance. To address this issue, we propose IPnP, an Iteratively Prompting and Pseudo-labeling framework, for partially labeled medical image segmentation. IPnP iteratively generates and refines pseudo-labels for unlabeled organs through collaboration between a trainable segmentation network (specialist) and a frozen foundation model (generalist), progressively recovering full-organ supervision. On the public dataset AMOS with the simulated partial-label setting, IPnP consistently improves segmentation performance over prior methods and approaches the performance of the fully labeled reference. We further evaluate on a private, partially labeled dataset of 210 head-and-neck cancer patients and demonstrate our effectiveness in real-world clinical settings.

cs.CV↗

arXiv2025

MacLaurin and Morality

Scottish mathematician Colin MacLaurin (1698-1746) is best known for his A Treatise of Fluxions (1742), An Account of Sir Isaac Newton's Philosophical Discoveries (1748), and the appellation for a type of power series. However, it is hardly known that in 1714 at the age of sixteen MacLaurin penned a short manuscript wherein he tried to apply Newtonian principles to morality, in an approach to mathematization that suggests strong continuities with earlier centuries. De viribus mentium bonipetis (On the good-seeking forces of minds) remained unpublished and hidden in the papers of the Colin Campbell Collection at the University of Edinburgh for over 250 years; it was only uncovered at the end of the twentieth century. De viribus provides a remarkable glimpse into how the young MacLaurin dealt with early Newtonianism, the tenets of the Church of Scotland, and the nascent interface between science and religion just prior to the dawn of the Scottish Enlightenment. Perhaps the most intriguing aspect of De viribus is the personal snippets related to Scottish Presbyterian morality that MacLaurin interjects throughout his mathematical discussion. These are often vague and oblique, and one must look to his mathematics, his contemporaries, and the social fabric of his surroundings to understand them. In the process, one gains insight not only into MacLaurin's family background and personal religious thought, but also into the culture and nature of teaching in Scottish universities at the time. De Viribus not only demonstrates that ideas of mathematising morality were being canvassed in Scotland much earlier than the better known, but less sophisticated, later attempts by the likes of Frances Hutcheson, David Hume and George Turnbull, but suggests continuities with approaches to mathematization in earlier centuries, providing a strong bridge into the Enlightenment era.

physics.hist-ph↗

arXiv2025

Federated prediction for scalable and privacy-preserved knowledge-based planning in radiotherapy

Background: Deep learning has potential to improve the efficiency and consistency of radiation therapy planning, but clinical adoption is hindered by the limited model generalizability due to data scarcity and heterogeneity among institutions. Although aggregating data from different institutions could alleviate this problem, data sharing is a practical challenge due to concerns about patient data privacy and other technical obstacles. Purpose: This work aims to address this dilemma by developing FedKBP+, a comprehensive federated learning (FL) platform for predictive tasks in real-world applications in radiotherapy treatment planning. Methods: We implemented a unified communication stack based on Google Remote Procedure Call (gRPC) to support communication between participants whether located on the same workstation or distributed across multiple workstations. In addition to supporting the centralized FL strategies commonly available in existing open-source frameworks, FedKBP+ also provides a fully decentralized FL model where participants directly exchange model weights to each other through Peer-to-Peer communication. We evaluated FedKBP+ on three predictive tasks using scale-attention network (SA-Net) as the predictive model. Conclusions: Our results demonstrate that FedKBP+ is highly effective, efficient and robust, showing great potential as a federated learning platform for radiation therapy.

cs.DC↗