arXiv subjects

Susanne Schulz-Heise

Publications and source records attributed to Susanne Schulz-Heise.

3 recordsLinked to original sources

arXiv2026

Panda: Unsupervised Pelvic Anomaly Detection for Real-Time MR Imaging

Female pelvic diseases remain an under researched area characterized by often delayed diagnosis. While pelvic MRI offers superior soft-tissue contrast for diagnosis and image-guided procedures, real-time anomaly detection remains challenging due to physiological motion, tissue deformation, and instrument artifacts. Existing supervised approaches are impractical, as adverse events are rare, heterogeneous, and difficult to annotate. We present a Dinomaly-based unsupervised anomaly detection framework adapted for pelvic MRI that learns normative representations from healthy cases and flags deviations without requiring labels. Our approach leverages a frozen DINOv3 Vision Transformer encoder combined with a noisy MLP bottleneck and Linear Attention decoder to prevent identity mapping while maintaining computational efficiency. Anomalies are localized via per-token cosine distance between encoder and decoder representations, yielding spatial anomaly maps that provide immediate feedback at the scanner to support radiologist decision-making and adaptive protocol adjustment. Evaluated on a curated subset of the Uterine Myoma Dataset, the framework achieves a pixel-level AUROC of 88.06% and high specificity (95.45%) at frame level at 40.5 slices/s, meeting real-time clinical deployment requirements. The spatial anomaly maps and frame-level scores provide immediate, localized feedback at the scanner to support radiologist decision-making and adaptive protocol adjustment during active procedures.

cs.CV↗

arXiv2026

Female-RHINO: A Real-Time Scanner-Integrated Framework for Automated Quantitative Uterine MRI Analysis and Structured Reporting

Standardized assessment of uterine MRI remains challenging due to anatomical variability, observer dependence, and the lack of workflow-integrated automated analysis tools. This work presents Female-RHINO: (R)eproductive (H)ealth (I)maging A(N)alysis T(O)ol, a real-time AI-assisted framework for automated quantitative uterine MRI analysis and structured reporting during image acquisition. We present an end-to-end system that integrates inline communication with the MRI scanner and deep learning-based analysis to derive quantitative uterine biomarkers from sagittal T2-weighted pelvic MRI. The framework combines segmentation and anatomical landmark detection models trained and evaluated on more than 500 multi-center datasets spanning diverse protocols, vendors, and patient populations. It performs volumetry, detects and quantifies common incidental findings such as fibroids and Nabothian cysts, and extracts six anatomical landmarks for biometric assessment. Results are compiled into a structured clinician-oriented report with integrated visualizations, without manual interaction. Evaluation on independent retrospective and prospective cohorts demonstrated robust performance across varying acquisition settings. Mean Dice similarity coefficients were 0.82 for the uterus and 0.80 for fibroids, with lower but consistent agreement for Nabothian cysts. Landmark detection achieved a mean radial error of 3.7 mm. End-to-end processing was completed in under 70 seconds, enabling availability of results during the ongoing scan. Prospective deployment yielded immediate, standardized, and reproducible analyses supported by inter-observer agreement. The proposed system enables real-time scanner-integrated AI for automated uterine MRI analysis and reporting, with potential to improve standardization, efficiency, and clinical workflow in pelvic imaging.

eess.IV↗

arXiv2025

Real-time MRI-based fetal femur length measurement

Purpose: To develop and evaluate a real-time method for automatic planning and measurement of fetal femur length - an important indicator of antenatal growth - during MRI. While routinely assessed by ultrasound, MRI-based femur length measurements remain challenging due to bone-slice misalignment, fetal motion, and the need for manual assessment. Methods: A low-latency 3D U-Net was trained on 59 scans acquired at 0.55T (gestational age 18-40 weeks) to localise the proximal and distal endpoints of the femur. These coordinates were employed to automatically adapt a 30-second EPI sequence for in-plane femur coverage in real-time. Retrospective evaluation was performed in 72 scans (19-39 weeks, including 19 pathological cases), and real-time testing in 24 cases (17-39 weeks). Automated results were compared with manual expert annotations and, in 57/72 cases, with matched clinical ultrasound measurements. Precision was evaluated against inter-observer variability and analysed across gestational age and maternal BMI. Furthermore, bilateral femur length consistency was evaluated. Results: Retrospective analysis demonstrated a mean endpoint localisation error of 5.5 mm and femur length deviation of 3.0 mm. Among the 49/72 cases with both femora automatically extracted, the left-right difference was 2.8+-2.7 mm. Precision was unaffected by maternal BMI (p>0.05), but correlated with gestational age (p<0.05). Real-time planning and assessment were successful in 22/24 cases, with a mean deviation of 4.2 mm. Conclusions: A real-time, automated method for MRI-based femur length measurement was developed, enabling precise, motion-robust, and operator-independent growth assessment, supporting future large-scale evaluation in growth-compromised pregnancies.

physics.med-ph↗