Aug 2026· Journal of Applied Clinical Medical Physics· Vol 27· 0 citations
Medicine
TL;DR
RatoGuide demonstrated favorable performance in typical cases, but accuracy declined in atypical cases with artifacts or altered anatomy, particularly for atypical cases and organs in high-dose gradient regions.
Abstract
Abstract Background Accurate contouring of target volumes and organs at risk is critical in radiotherapy. While deep learning (DL) models offer automated contouring, their clinical applicability to real‐world cases containing anatomical variations and artifacts requires rigorous validation. Purpose To evaluate the clinical accuracy and potential vulnerabilities of RatoGuide, novel DL‐based auto‐segmentation software, using a dataset including atypical cases derived from routine clinical practice. Methods This single‐center retrospective study included 69 thoracic and male pelvic cases. The cohort was intentionally selected to encompass diverse anatomies and artifacts (e.g., pacemakers, SpaceOAR implants, artificial femoral head replacements, and unilateral atelectasis). Auto‐contours generated by RatoGuide were compared with expert‐approved manual contours. Performance was evaluated quantitatively using the Dice Similarity Coefficient (DSC) and 95th percentile Hausdorff Distance (HD95), and qualitatively via a 5‐point visual assessment scale by four independent reviewers. Statistical comparisons between cohorts were performed using the Mann‐Whitney U test. Additionally, a dosimetric evaluation was conducted for male pelvic cases to assess clinical impact. Results In typical cases, the software maintained high segmentation accuracy (thorax: mean DSC 0.856, mean HD95 6.89 mm; male pelvis: mean DSC 0.874, mean HD95 4.11 mm). However, performance declined in atypical cohorts (thorax: mean DSC 0.808, p = 0.0457, mean HD95 12.22 mm, p = 0.0002; male pelvis: mean DSC 0.828, p = 0.1620, mean HD95 6.16 mm, p = 0.0075). Notable decreases in accuracy were observed in challenging scenarios, such as artificial femoral head replacements (DSC: 0.754) and unilateral atelectasis (DSC: 0.784). Qualitative assessment revealed that errors were primarily due to anatomical factors and artifacts. Furthermore, the dosimetric evaluation identified one critical false‐negative error where a dose constraint violation was overlooked when the DL contour was used. Conclusions RatoGuide demonstrated favorable performance in typical cases, but accuracy declined in atypical cases with artifacts or altered anatomy. For clinical implementation, rigorous visual verification and manual review by experts are essential, particularly for atypical cases and organs in high‐dose gradient regions.
MAGE explains how externalized knowledge, bounded action, independent evaluation, and retained human authority can compose into a governed engineering environment, and proposes tests of when that environment turns commodity intelligence into durable engineering progress.
James C. Davis, Kelechi G. Kalu, Huiyun Peng et al.· 1 citation
LLMs are increasingly used for code generation, yet they frequently hallucinate non-existent software packages, creating exploitable entry points into the software supply chain. We make four contributions to this problem. First, we show that prior evaluation methodologies systematically inflate hallucination rates by misclassifying standard-library modules as hallucinations in some languages. For Python, the overestimation reaches 9.4 percentage points. Second, we evaluate seven inference-time defenses for mitigating package hallucinations, including five guided decoding strategies (Greedy, Contrastive, DoLa, Nudging, and Active Layer-Contrastive Decoding), an iterative self-refinement approach (Self-Refine), and a Retrieval-Augmented Generation (RAG)-based defense.. Across eight models spanning five families and four programming languages (Python, JavaScript, Ruby, Rust), RAG reduces the package hallucination rate (PHR) in 18 of 32 model--language configurations. Third, we introduce Package Utility (PU) to assess whether defenses preserve valid and task-relevant recommendations. Among strategies evaluated, Greedy decoding provides the strongest average mitigation--utility trade-off. Fourth, we stress-test all strategies under adversarial prompts seeded with fabricated package names and find that PHR surges by up to 45 percentage points relative to standard prompts, with Ruby consistently the most vulnerable language (80.9--95.2\%). Under adversarial conditions, RAG and Self-Refine outperform all decoding-only strategies, indicating that robust defense requires either external grounding or iterative self-verification when prompts are actively hostile. Our results recast package hallucination as both a measurement problem and a decoding-time control problem, and they demonstrate that the choice of defense must be matched to the threat model and recommendation utility.
Albérick Euraste Djiré, Iyiola E. Olatunji, Melissa Tessa et al.· 1 citation
Post-training alignment is often shallow, eroding under fine-tuning. It remains untested as to whether constitutional midtraining interventions can produce durable alignment when cleanly isolated from post-training. We build a 394M-token constitutional corpus from Anthropic's Constitution and apply constitutional midtraining at 120B scale, where principled, values-based content is inserted into midtraining. A 2x2 design (curriculum ordering x deliberative reasoning) was used to produce four constitutionally midtrained conditions, plus a control, which were evaluated on self-generated and established benchmarks including alignment under pressure, value conflict resolution, blackmail, and emergent misalignment. All models were evaluated across three stages: post-midtraining, post-SFT, and post-benign fine-tuning. Constitutionally midtrained models outperformed the control on alignment generalization and durability, notably on blackmail: SFT instilled a blackmail propensity in all models, but constitutional midtraining blunted it, with the advantage surviving benign fine-tuning (-17.5pp). This durability did not extend to settings that required active resistance to in-context pressure or conflict, where the advantage attenuates after SFT. The presence of constitutional content at midtraining also mattered more than its structure, and constitutional midtraining incurred no capability cost, on average, at any stage (MMLU, ARC-Easy, piqa, GSM8K). A modest amount of constitutional content at midtraining could therefore yield broad, persistent alignment gains, offering a cheap, complementary addition to SFT-centered pipelines. Code, data, and models are available.
Desiree Cho, Cameron Tice, Bernie Hogan et al.· 0 citations
Lower gestational age and lower birth weight were significantly associated with persistence of PDA, indicating prematurity as a major determinant, highlighting PDA as a prevalent condition contributing significantly to early complications and death.
Avinash Shukla, Amar M Taksande· Journal of Clinical and Diag...· 0 citations
The findings of the present study indicated that potential complications such as delayed union, nonunion, and osteomyelitis in the intramedullary nailing method are approximately comparable to those of the external fixator method.
Reza Noktesanj, Ali Nami, F. Amani et al.· journal of Health Research a...· 0 citations
The positive results of ISIT counseling on psychological distress symptoms and reducing perinatal grief during follow-ups conducted immediately post-intervention and 3 months after the intervention suggest that it will be useful for managing the complications of spontaneous abortion.
Nazanin Karimihamzekolaee, Hajar Adib-rad, Hajar Pasha et al.· Health Science Reports· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduAug 17, 2026
A USAF cadet and a Lincoln Laboratory researcher found AI chatbots can help nontechnical service members produce viable software applications for their unique problems.
