4376 papers
This large retrospective cohort study of over 384,000 pregnancies found that several pre-pregnancy chronic health conditions are significantly associated with an increased risk of spontaneous abortion, leading to the development of an exploratory risk score that showed moderate discrimination but requires further validation before clinical application.
This study demonstrates that two-week postoperative T10 tilt is an independent prognostic marker for proximal junctional kyphosis and failure in adult spinal deformity patients with an upper instrumented vertebra at T9 or T10, offering modest predictive value comparable to or slightly better than L1 pelvic angle for early risk assessment.
This study utilizes single-cell and spatial profiling to map the heterogeneous distribution and activity of cysteine cathepsins across breast cancer ecosystems, revealing that integrating protease abundance and activity with target antigen expression is crucial for optimizing the design and patient stratification of protease-cleavable antibody-drug conjugates.
This randomized controlled trial demonstrates that Juniver, a digital intervention combining neuroscience, CBT, AI support, and moderated peer groups, is significantly more effective and safe than a wait-list control in reducing eating disorder symptoms, depression, anxiety, and psychosocial impairment among adults.
This study demonstrates that low-pass whole-genome sequencing-based copy number variation profiling of pleural effusion cfDNA is a highly sensitive, noninvasive, and cost-effective tool for accurately diagnosing lung adenocarcinoma-associated malignant pleural effusion and stratifying patient prognosis, outperforming traditional cytology and CEA testing.
In a real-world study of locally advanced cervical cancer patients undergoing platinum-taxane-based concurrent chemoradiotherapy, prophylactic PEG-rhG-CSF significantly reduced severe neutropenic events and improved chemotherapy delivery compared to reactive short-acting support.
This study demonstrates that machine learning models integrating clinical and pharmacokinetic variables outperform traditional methods in predicting busulfan exposure and survival, thereby enabling more precise, personalized dosing strategies for patients undergoing hematopoietic stem cell transplantation.
This proof-of-concept study demonstrates that a high-intensity theatre model utilizing parallel anaesthetic preparation enables the safe and efficient completion of fifteen major robot-assisted thoracic resections with 100% list completion and zero mortality, significantly increasing surgical throughput without additional resources.
This study developed and validated an interpretable multimodal deep learning framework that integrates biparametric MRI sequences with clinical parameters to achieve highly accurate, objective stratification of early- versus late-stage osteonecrosis of the femoral head, thereby aiding critical decision-making for hip preservation.
This monocentric pilot study demonstrates that merging routine data across emergency, inpatient, and intensive care sectors is feasible and effectively reveals heterogeneities in sepsis care processes, thereby identifying critical areas for optimization to improve patient outcomes.