Search

Objective: This study introduces an innovative end-to-end deep learning pipeline designed to automatically classify and order fetal ultrasound standard planes in alignment with the guidelines of the Canadian Association of Radiologists, while also assessing the diagnostic usability of each view. The primary objective is to address the manual and cumbersome challenges that interpreting radiologists encounter in the existing obstetric ultrasound workflow. Methods: We compiled a diverse dataset, comprising 33,561 de-identified two-dimensional obstetrical ultrasound images acquired from January 1, 2010, to June 1, 2020. This dataset was categorized into 19 distinct classes associated with standard planes and further partitioned into training, validation, and testing subsets via a 60:20:20 stratified split. The standard plane and diagnostic usability networks are founded on a convolutional neural network framework and employ the benefits of transfer learning. Results: The standard plane classification network demonstrated promising results by achieving 99.4 % and 98.7 % for accuracy and F1 score, respectively. Subsequently, the diagnostic usability network demonstrated strong performance, registering 80 % accuracy and an 82 % F1 score. Notably, this study is the first to investigate whether deep learning methods can surpass sonographers in the standard plane labeling task, with some instances revealing the algorithm's capacity to rectify sonographer mislabeled planes. Conclusion: The results highlight the algorithm's potential to be integrated into a clinical setting by serving as a reliable assistive tool, alleviating the cognitive workload faced by radiologists and enhancing efficiency and diagnostic outcomes in the current obstetric ultrasound process.

Abstract While it is common to monitor deployed clinical artificial intelligence (AI) models for performance degradation, it is less common for the input data to be monitored for data drift – systemic changes to input distributions. However, when real-time evaluation may not be practical (eg., labeling costs) or when gold-labels are automatically generated, we argue that tracking data drift becomes a vital addition for AI deployments. In this work, we perform empirical experiments on real-world medical imaging to evaluate three data drift detection methods’ ability to detect data drift caused (a) naturally (emergence of COVID-19 in X-rays) and (b) synthetically. We find that monitoring performance alone is not a good proxy for detecting data drift and that drift-detection heavily depends on sample size and patient features. Our work discusses the need and utility of data drift detection in various scenarios and highlights gaps in knowledge for the practical application of existing methods.

Abstract The overuse of low value medical tests and treatments drives costs and patient harm. Efforts to address overuse, such as Choosing Wisely campaigns, typically rely on passive implementation strategies- a form of low reliability system change. Embedding guidelines into clinical decision support (CDS) software is a higher leverage approach to provide ordering suggestions through an interface embedded within the clinical workflow. Growth in computing power is increasingly enabling artificial intelligence (AI) to augment such decision making tools. This article offers a roadmap of opportunities for AI-enabled CDS to reduce overuse, which are presented according to a patient’s journey of care.

‎Groups of $F$-type were introduced in [B‎. ‎Fine and G‎. ‎Rosenberger‎, ‎Generalizing Algebraic Properties of Fuchsian Groups‎, \emph{London Math. Soc. Lecture Note Ser.}, \textbf{159} (1991)‎ ‎124--147.] as a natural algebraic generalization of Fuchsian groups‎. ‎They can be considered as the analogs of cyclically pinched one-relator groups where torsion is allowed‎. ‎Using the methods In [B‎. ‎Fine‎. ‎M‎. ‎Kreuzer and G‎. ‎Rosenberger‎, ‎Faithful Real Representations of Cyclically Pinched One-Relator Groups, Int. J. Group Theory, 3 ‎(2014) 1--8.] we prove that any hyperbolic group of $F$-type has a faithful representation in $PSL(2,\mathbb R)$‎. ‎From this we also obtain that a cyclically pinched one-relator group has a faithful real representation if and only if it is hyperbolic‎. ‎We further survey the many nice properties of groups of $F$-type‎. ‎

In [FGRS1,FGRS2] the relationship between the universal and elementary theory of a group ring $R[G]$ and the corresponding universal and elementary theory of the associated group $G$ and ring $R$ was examined. Here we assume that $R$ is a commutative ring with identity $1 \ne 0$. Of course, these are relative to an appropriate logical language $L_0,L_1,L_2$ for groups, rings and group rings respectively. Axiom systems for these were provided in [FGRS1]. In [FGRS1] it was proved that if $R[G]$ is elementarily equivalent to $S[H]$ with respect to $L_{2}$, then simultaneously the group $G$ is elementarily equivalent to the group $H$ with respect to $L_{0}$, and the ring $R$ is elementarily equivalent to the ring $S$ with respect to $L_{1}$. We then let $F$ be a rank $2$ free group and $\mathbb{Z}$ be the ring of integers. Examining the universal theory of the free group ring ${\mathbb Z}[F]$ the hazy conjecture was made that the universal sentences true in ${\mathbb Z}[F]$ are precisely the universal sentences true in $F$ modified appropriately for group ring theory and the converse that the universal sentences true in $F$ are the universal sentences true in ${\mathbb Z}[F]$ modified appropriately for group theory. In this paper we show this conjecture to be true in terms of axiom systems for ${\mathbb Z}[F]$.

The group Γ = PSL(2, C) arises in a wide variety of contexts; hyperbolic geometry, automorphic function theory, number theory and group theory. Much of combinatorial group theory arose out of the study of discrete subgroups of Γ = PSL(2, C), in particular Fuchsian Groups and Kleinian groups. From the Poincaré polygon theorem surface groups can be faithfully represented in PSL(2, C). Extending this, most cyclically pinched one-relator groups can also be embedded in Γ . Recent results of Fine and Rosenberger ([61],[62]) show that all finitely generated fully residually free groups, the so called limit groups, can also be faithfully represented in this group. In this paper we survey the tremendous impact this single group has had on combinatorial group theory in particular and infinite group theory in general.