The study introduces Spatial Atlas-Informed Loss Ensembles (SAILS) for whole-body PSMA PET/CT metastasis segmentation, integrating a population metastasis atlas into nnU-Net training via voxel-wise mean and standard deviation maps. SAILS improves lesion detection, notably in high-risk bone regions, achieving higher Dice scores and sensitivity on AutoPET~III and Deep-PSMA datasets compared to baseline methods, though it also increases false-positive volume. The approach eliminates the need for atlas registration at inference time.
arXiv:2609.24769v1 Announce Type: new Abstract: Brain metastases are the most common intracranial malignancy, occurring in roughly 30% of patients with primary solid tumors and carrying a median surv...
The study evaluates whether overlap metrics like Dice can reliably indicate metabolic tumor volume (MTV) estimation accuracy in PET segmentation models. Across five FDG PET/CT models, higher Dice scores correlated with lower relative MTV error on average, but at the patient level, 16.8% of cases fell on opposite sides of clinically relevant thresholds. Additionally, many models missed lesions, showing that model ranking alone does not guarantee reliable individual MTV estimates.
SADL is a size‑adaptive decision layer that improves chronic stroke lesion segmentation on T1‑weighted MRI by combining an eleven‑member nnU‑Net ensemble with adaptive thresholding and pruning. The ensemble incorporates dice + top‑K loss, heavy augmentation, per‑lesion volume re‑weighting, stochastic weight averaging, and a cohort‑balanced sampling block with a residual‑encoder backbone. On a 91‑case holdout, SADL achieves Dice 0.7322, lesion‑F1 0.7461, and PR‑AUC 0.8430, with a statistically significant gain in lesion‑F1 (p=0.024) especially for small lesions.
The paper reports the submissions of team nucli to the HECKTOR 2026 challenge, which evaluates machine learning methods for head‑and‑neck cancer imaging. The team used self‑supervised pre‑trained encoders, training a 2D DINOv2 encoder on PET scans and combining a DynUNet with a SwinUNETR for segmentation, achieving a mean Dice of 0.5409. For TN staging, they fused geometric features from segmentation masks with PET‑DINOv2‑MIPs deep features via logistic regression, obtaining balanced accuracies of 0.3962 (T‑stage) and 0.6432 (N‑stage). For recurrence‑free survival prognosis, they fine‑tuned the PET‑DINOv2‑MIPs ViT‑S encoder with an MLP head under a CoxPH loss, achieving a Harrell's C‑index of 0.6577 and a weighted challenge score of 0.5802.
The paper introduces FORESEE-MS, a protocol for evaluating prospective spatial forecasting of future multiple sclerosis lesions using baseline FLAIR MRI. It compares a fold‑specific population‑prior atlas to five deep‑learning models (3D U‑Net, Attn+ResBlock+VAE, ViT, and two SwinUNETR variants) on voxel‑level discrimination (AUROC, AUPRC) and within‑patient lesion localization (Recall@Top‑K%). While the atlas achieved the best pooled AUROC and AUPRC, the 3D U‑Net and Attn+ResBlock+VAE showed higher within‑patient recall, though confidence intervals largely overlapped zero, leaving the benefit of patient‑specific predictive information uncertain in this small cohort.
The paper presents a segmentation pipeline for brain metastases in both pre‑ and post‑treatment cases using a 5‑fold nnU‑Net ResEnc‑L ensemble trained on 1,296 four‑modality cases. A rule‑based post‑processing cascade improves the lesion‑wise Dice similarity coefficient (LW‑DSC) for enhancing tumour, tumour core, whole tumour, and resection cavity sub‑regions, achieving LW‑DSC scores of 0.733, 0.751, 0.713, and 0.549 respectively on the official validation leaderboard. The authors conduct a five‑fold out‑of‑fold analysis to validate the robustness of each post‑processing stage, provide a mechanistic explanation of LW‑DSC behaviour, and report thirteen negative results that challenge common intuitions, with all code released under Apache‑2.0.
The paper presents a pragmatic segmentation pipeline for brain metastases in the BraTS 2026 Task 1, using a 5‑fold nnU-Net ResEnc‑L ensemble trained for 1,000 epochs on 1,296 four‑modality cases. A rule‑based post‑processing cascade tuned for the lesion‑wise Dice similarity coefficient (LW‑DSC) improves performance, achieving LW‑DSC scores of 0.733, 0.751, 0.713, and 0.549 on enhancing tumour, tumour core, whole tumour, and resection cavity, respectively. The authors audit each post‑processing stage with a five‑fold out‑of‑fold analysis, confirm two stages as robust, and provide a mechanistic analysis of LW‑DSC, along with thirteen negative results that challenge common intuitions.
arXiv:2609.13838v1 Announce Type: new Abstract: Purpose: To develop and evaluate TotalSynth, a reusable pretrained framework for whole-body synthetic CT (sCT) generation from MRI and cone-beam CT (CB...
SPARCL (Spatially Adaptive Regional Coverage for Active Learning) is a new active learning strategy for chest X‑ray lesion segmentation that balances predictive uncertainty with spatial coverage using a concave coverage function. It directs annotation effort toward under‑labeled regions, reverts to standard uncertainty sampling when spatial variation is absent, and guarantees near‑optimal greedy batch selection through a monotone submodular objective. The authors also propose Spatial Inconsistency and Performance Coverage metrics to reveal regional fairness, and demonstrate that SPARCL achieves 95% of fully supervised performance using only 24% of the data versus 29% for the best baseline.
In clinical oncology studies, metastatic cancer is commonly evaluated using "Response Evaluation Criteria in Solid Tumors" (RECIST), in which the diameter of up to five lesions is measured and followed over the course of treatment. However, RECIST shows limited correlation with overall survival.
The paper introduces TAP-JEPA, a joint‑embedding predictive architecture trained on over 113,000 CT volumes, which achieves performance comparable to leading encoders in scan‑level classification. It demonstrates a predictability bias: while anatomy is largely predictable from context, pathology is not, leading to significant degradation in pathology detection when predictions are applied. The study audits the model’s predictive pathway, showing that pathology representations shift toward control regions and that removing pathology tokens minimally affects downstream decisions, underscoring that contextual predictability alone is insufficient for clinical relevance.
arXiv:2606. 28392v1 Announce Type: cross Abstract: Accurate lesion segmentation in PET/CT is critical for oncology, yet remains challenging because physiologic tracer uptake and artifacts can mimic malignant signal.