Search Results (1,940)

Computational fluid dynamics (CFD)-based simulation of coronary blood flow provides valuable hemodynamic markers, such as pressure gradients, for diagnosing coronary artery disease (CAD). However, CFD is computationally expensive, time-consuming, and difficult to integrate into large-scale clinical workflows. These limitations restrict the availability of labeled hemodynamic data for training AI models and hinder the broad adoption of non-invasive, physiology-based CAD assessment. To address these challenges, we develop an end-to-end pipeline that automates coronary geometry extraction from coronary computed tomography angiography (CCTA), streamlines simulation data generation, and enables efficient learning of coronary blood pressure distributions. The pipeline reduces the manual burden associated with traditional CFD workflows while producing consistent training data. Furthermore, we introduce a diffusion-based regression model. Specifically, the inverted conditional diffusion (ICD) model is designed to predict coronary blood pressure directly from CCTA-derived features, thereby bypassing the need for computationally intensive CFD during inference. The proposed model is trained and validated on two CCTA datasets using the Adam optimizer with a weight decay of $1\times {10}^{-3}$, a learning rate of $1\times {10}^{-5}$, a batch size of 100, and Huber loss. It is then evaluated on a test set of ten simulated coronary hemodynamic cases. Experimental results demonstrate state-of-the-art performance. Compared with Long Short-Term Memory (LSTM), the proposed model improves the $R^2$ score by 19.78%, reduces the root mean squared error (RMSE) by 19.44%, and lowers the normalized root mean squared error (NRMSE) by 18%. Compared with a multilayer perceptron (MLP), it improves the $R^2$ score by 8.38%, reduces RMSE by 4.3%, and reduces NRMSE by 5.4%. This work represents a first step toward a scalable and accessible framework for rapid, non-invasive, CFD-based blood pressure prediction, with the potential to support CAD diagnosis. Full article

The overexpression of hypoxia-inducible factor-1α (HIF-1α) suppresses STING signaling and modulates lipid metabolism in tumor cells, leading to abnormal lipid droplet (LD) accumulation. Herein, we constructed a manganese dioxide (MnO₂)-based nanomotor (HMIP@A). HMIP@A depletes intracellular hydrogen peroxide (H₂O₂) and glutathione (GSH) to generate oxygen (O₂), reactive oxygen species (ROS), and manganese (Mn²⁺). A dual strategy of “oxygen supplementation” and “small-molecule inhibition” synergistically downregulates HIF-1α, thereby suppressing LD biogenesis. This process sensitizes tumor cells to ROS, leading to severe DNA damage. Released Mn²⁺ and damaged DNA synergistically activate the cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) pathway. In vitro, HMIP@A markedly increases ROS production, lipid peroxidation (LPO), and DNA damage, thereby inducing tumor cell death, immunogenic cell death (ICD), and dendritic cell (DC) maturation. Furthermore, HMIP@A exhibits excellent penetration in tumor spheroids. Overall, this study provides a theoretical basis for the design of nanomedicines through a strategy integrating metabolic intervention, oxidative damage sensitization, and immune activation. Full article

This systematic review examines the emerging interplay between ferroptosis and disulfidptosis, two distinct forms of regulated cell death (RCD) centered on the SLC7A11 (also known as xCT)-mediated metabolic paradox. Traditionally recognized as a potent anti-ferroptotic factor, SLC7A11 imports cystine for glutathione synthesis to neutralize iron-dependent lipid peroxidation. However, the discovery of disulfidptosis identifies SLC7A11 as a metabolic liability, representing a paradigm shift in our understanding of cellular antioxidant defense. This discovery reveals a transformative vulnerability in SLC7A11-overexpressing cells, shifting the focus from conventional survival mechanisms to the consequences of catastrophic structural collapse. Beyond metabolic exhaustion, this review highlights the role of metal dyshomeostasis as a primary driver, spanning from iron-catalyzed ferroptosis to copper-mediated metabolic interference. This conceptual framework redefines the SLC7A11 axis as a targetable “double-edged sword” in therapy-resistant malignancies. Clinical synthesis of multi-omic gene signatures, such as the disulfidptosis- and ferroptosis-related gene prognostic score (DRGPS) and the ferroptosis- and disulfidptosis-related gene (FDRG) scores, demonstrates their robust value in prognostic stratification and in predicting immunotherapy response across malignancies, including lung adenocarcinoma and hepatocellular carcinoma. Furthermore, we evaluate the capacity of disulfidptosis to prime immunogenic cell death (ICD) and remodel the immunosuppressive tumor microenvironment to bypass chemoresistance. By integrating mechanistic insights with clinical data, this review provides a comprehensive framework for targeting the SLC7A11 axis as a transformative therapeutic vulnerability in precision oncology. Full article

Background/Objectives: Excessive trabeculation of the left ventricle, previously known as left ventricular non-compaction (LVNC), is a rare phenotypic trait whose mechanisms and pathogenesis still remain conflictual. Its presentations may range from heart failure to embolism and, most importantly, ventricular arrhythmias (VAs). This study aims to find novel predictive factors for the occurrence of potentially fatal VAs in patients with left ventricular hypertrabeculation. Methods: All consecutive patients meeting the echocardiographic (Chin, Jenny or Stöllberger) and/or MRI criteria (Petersen) for hypertrabeculation were prospectively enrolled from October 2009 to December 2023. The primary outcome was a composite of sudden cardiac death, sustained ventricular tachycardias (sVTs), ventricular fibrillation (VF) or appropriate implantable cardioverter defibrillator (ICD) interventions. The secondary outcome was a composite of cardiovascular death and cardiovascular hospitalizations. Results: Overall, 64 patients (41 males, mean age 46 ± 19 years old) were enrolled and followed for a median time of 2.2 years. Six patients (9.4%) experienced a composite outcome at eight years, three with previous sVTs and three with previous non-sustained VTs (nsVTs). The strongest predictor of the primary endpoint was the anamnesis of nsVTs and sVTs before LVNC diagnosis. In addition, nsVTs and sVTs were significantly associated with the secondary outcome. Conclusions: Hypertrabeculation of the left ventricle is a complex and poorly understood condition whose status of cardiomyopathy is currently challenged. In our population, patients with a trabecular pattern experienced a high incidence of VAs, cardiovascular death and hospitalizations. VAs before LVNC diagnosis were predictive of the outcome independently from systolic function. Full article

Background: Modern oncology views immune system dysfunction as a key factor in carcinogenesis. The induction of immunogenic cell death (ICD), a form of regulated cell death capable of activating adaptive immunity, represents a promising therapeutic strategy. Damage-associated molecular patterns (DAMPs) play a central role in this process. This review aims to summarize current knowledge of DAMPs, their release mechanisms during ICD, their classification, and their prognostic and therapeutic significance in antitumor immunity. Methods: We systematically reviewed and synthesized literature published in Pubmed and Google Scholar on ICD and DAMPs, focusing on distinct forms of DAMPs which were categorized based on recognition mechanisms (five classes) and cellular origin (extracellular, mitochondrial, nuclear, and cytosolic). Key molecules, their receptors, downstream signaling pathways, and clinical associations were analyzed. Results: The spatiotemporally coordinated release of the pattern of DAMPs promotes dendritic cell maturation, antigen presentation, activation of cytotoxic T lymphocytes, and elimination of tumor cells. DAMPs can exhibit a dual role: they are able to induce sterile inflammation essential for antitumor immunity, but may also contribute to metastasis and chronic inflammation. Among all DAMPs, high-mobility group box 1 (HMGB1, a nuclear DAMP) and calreticulin (CRT, a cytosolic protein) demonstrate the greatest prognostic value. Other DAMPs (e.g., extracellular matrix components, uric acid) act as signal amplifiers during various forms of cell death. Conclusions: Understanding the spatiotemporal dynamics of DAMP release is critical for activating immune responses against malignant cells. Monitoring DAMPs may improve patient stratification, predict therapeutic responses, and enable personalized immunotherapeutic strategies. Further investigation of ICD mechanisms and DAMP release represents a fundamental basis for developing novel anticancer therapies. Full article

Laser surface hardening is a technique that improves various mechanical characteristics of different materials. The methods are being extensively used in the automobile, aerospace, tool manufacturing, and construction industries for various components. The present review highlights the hardness and hardened surface depth improvement of different steels and non-ferrous alloys in as-bought and pre-heat treatment conditions. Diode and fibre lasers have rendered higher surface hardness and hardened depth, while consuming higher power. Nd:YAG lasers have resulted in a precise increase in hardness and a very minimal 0.8 in ferrous and 2 mm in surface-hardened depth of non-ferrous alloys, proving a better efficiency. The pre-heat treatments are selected to enhance mechanical properties and reduce the deformations and defects. An increase of 300.43 and 282.38% of surface hardness due to laser hardening as compared to the core material of AISI 420 was observed using a high-power diode laser. A huge 281.41% of increase in surface hardness was observed for ICD-5 tool steel using Nd:YAG lasers. The annealing pre-heat treatment has also affected the hardenability, resulting in high hardness. Non-ferrous alloys such as titanium and A356 alloys have recorded 200 and 125% increase in surface hardness compared to their core using Nd:YAG lasers. Full article

Background: Bushfire smoke exposure (BFSE) is associated with adverse pregnancy and neonatal outcomes; however, the specific impact of BFSE on pregnancies complicated by asthma is not well characterised. Methods: A retrospective cohort study analysed data from 22,166 pregnant women who gave birth in the Illawarra Shoalhaven region between January 2017 and December 2022. Women with asthma were identified by the ICD-10-AM code for asthma during hospital admission for birth. Exposure was defined using a fixed time-window assumption. Women were considered exposed to bushfire smoke if they experienced at least 4 weeks of their pregnancy between 25 October 2019 and 4 February 2020. Results: Prevalence of asthma in the total population was 8.31%. In the control cohort, outcomes for pregnant women with asthma were poorer than those without. Pregnant women with BFSE had increased odds of postpartum haemorrhage (OR 1.603; 95% CI 1.42–1.81), and decreased odds of gestational hypertension (OR 0.615; 95% CI 0.49–0.77), gestational diabetes mellitus (OR 0.703; 95% CI 0.63–0.79) and preterm birth (OR 0.813; 95% CI 0.67–0.98). Maternal asthma did not confound the relationship between BFSE and any of the primary study outcomes. Conclusions: This study emphasises the independent effects of asthma on pregnancy outcomes. The impact of BFSE on pregnant women with asthma remains unclear. Further research is needed to characterise the true effect of BFSE on pregnancies, uncomplicated and complicated by asthma. Full article

Background: Gastric cancer remains one of the leading causes of cancer-related mortality worldwide, underscoring the need for more effective diagnostic strategies. This study aims to use annotated digitized histological slides of gastric cancer and precancerous lesions to develop artificial intelligence algorithms for the diagnosis of gastric lesions. Materials and Methods: We developed a deep learning tool using a training cohort of 970 digitized gastric biopsy slides. Convolutional neural networks (CNNs) were trained for histological recognition and ICD-10 code assignment. The model was validated on an independent test cohort of 250 cases, with expert consensus as the reference standard. Performance was assessed using sensitivity, specificity, and Cohen’s kappa. Survival analysis used Kaplan–Meier, log-rank tests (SPSS 16.0; p < 0.05 significant). Results: Analysis of the training cohort led to a scoring system predicting fatal outcomes based on age and morphology (high-grade component > 70%, ulceration, absence of metaplasia/dysplasia). High-risk patients (4–5 points) had significantly worse survival than low-risk patients (0–3 points) (Log Rank = 14,754; p < 0.0001). One-year survival was 71% (low-risk) vs. 40% (high-risk); mean survival was 19.2 vs. 11.3 months. In the test cohort, the AI algorithm demonstrated 79.6% sensitivity and 86.7% specificity (p < 0.0001) for differentiating malignant from benign gastric lesions. Conclusions: A system combining AI-based analysis with a prognostic scoring model has been developed to reduce diagnostic errors and improve risk stratification in gastric cancer pathology. Full article

Background/Objectives: Ischemic stroke, hemorrhagic stroke, and transient ischemic attack (TIA) differ in terms of medical severity and prognosis; however, it remains unclear whether these differences are reflected in nursing complexity and nursing intensity when assessed using standardized nursing language. Methods: This retrospective study analyzed routinely collected nursing and administrative data from an acute care hospital. Hospitalizations were classified as ischemic stroke, hemorrhagic stroke, or TIA using ICD-9-CM codes. Nursing complexity was measured as the number of nursing diagnoses (NDs) documented within 24 h of admission, while nursing intensity was measured as the number of nursing actions (NAs) recorded during hospitalization. Group differences were tested using ANOVA and Kruskal–Wallis tests, as appropriate. Results: A total of 728 hospitalizations were included: 429 ischemic strokes, 236 hemorrhagic strokes, and 63 TIAs. Overall, 4136 NDs and 27,528 NAs were recorded. Distinct patterns emerged across stroke categories. ND counts differed significantly (F = 5.81, p = 0.003), with TIA showing lower counts than both ischemic and hemorrhagic stroke, while no significant difference was observed between ischemic and hemorrhagic stroke. NA counts also differed significantly (H = 16.73, p < 0.001), with the highest counts in hemorrhagic stroke, intermediate counts in ischemic stroke, and the lowest counts in TIA. In a sensitivity analysis standardized by length of stay, nursing intensity also differed significantly across stroke categories (H = 12.999, p = 0.002), although the pattern differed from that observed for cumulative counts. Conclusions: Nursing complexity and nursing intensity showed distinct patterns across stroke categories. While complexity was comparable between ischemic and hemorrhagic stroke and lower in TIA, intensity followed a clear gradient, highest in hemorrhagic stroke, intermediate in ischemic stroke, and lowest in TIA. Standardized nursing data may complement medical indicators by capturing additional dimensions of patient needs and care delivery in people with stroke. Full article

Background: Dyspnea is a common reason for emergency medical service (EMS) interventions and is associated with a substantial risk of severe clinical course, complications, and hospital admission. Its differential diagnosis in the prehospital setting remains challenging due to the limited availability of imaging modalities. Point-of-care ultrasound (POCUS), including lung ultrasound (LUS), is a rapid, field-applicable technique recommended in numerous acute respiratory diagnostic scenarios. Objective: To evaluate the use of lung ultrasound in the prehospital setting and its association with the precision of diagnoses related to respiratory failure, the frequency of transport to the emergency department (ED) among patients presenting with dyspnea/respiratory failure, and to characterize the profile of sonographic findings with their correlation to clinical diagnostic categories. Additionally, transport rates in the study population were compared with aggregated regional data for the Masovian Voivodeship (excluding the analyzed county). Methods: A retrospective observational study was conducted on EMS interventions performed between 01 January 2025 and 30 June 2025 in Legionowo County (N = 353). The analysis included ICD-10 codes assigned in prehospital documentation (one primary code and up to two additional codes) in patients presenting with dyspnea and/or respiratory failure, the performance of ultrasound examination, and resulting LUS findings (absence of pleural sliding and/or lung point; B-lines; consolidations; C-lines; pleural effusion). Descriptive analyses, frequency comparison tests (χ²/Fisher), estimation of relative risk (RR) with 95% confidence intervals (CI), and agreement analysis using Cohen’s kappa coefficient (κ) between etiological categories derived from ICD-10 codes and those inferred from LUS profiles were performed (κ with 95% CI estimated using bootstrap resampling). The study was reported in accordance with the STROBE guidelines for observational studies. Additionally, the distribution of ICD-10 coding and the proportion of hospital transports across the entire Masovian Voivodeship were compared with those observed in the analyzed area. Results: Ultrasound examination was performed in 72/353 (20.4%) EMS interventions; transport to the emergency department occurred in 239/353 (67.7%) cases. The most frequent clinical categories based on ICD-10 codes were: general/symptom-based 182/353 (51.6%), inflammatory 77/353 (21.8%), obstructive 66/353 (18.7%), and cardiological 20/353 (5.7%). Among abnormal LUS findings, the most common were B-lines (43/72; 61.4%) and consolidations (29/72; 41.4%). Consolidations were strongly associated with the inflammatory category (OR 9.72; p < 0.001), whereas B-lines were associated with the cardiological category (OR 23.41; p = 0.0011) among cases in which LUS was performed. Ultrasound use was associated with a higher frequency of assigning at least one targeted (non-symptom-based) diagnosis within ICD coding: 53/72 (73.6%) vs. 111/278 (39.9%), RR 1.84 (95% CI 1.51–2.25; p < 0.001). Agreement between the ICD-10 etiological category (inflammatory/cardiological/obstructive/other) and the category inferred from the LUS profile was moderate: κ = 0.36 (95% CI 0.21–0.51), with an observed agreement of 54.2%. Compared with aggregated regional data (Masovian Voivodeship excluding the analyzed county), the overall transport rate for comparable ICD-10 codes was lower in the study unit: 279/409 (68.2%) vs. 11,351/13,785 (82.3%), RR 0.83 (95% CI 0.78–0.89; p < 0.001). The largest differences were observed for dyspnea (R06.0: 72.9% vs. 88.2%; RR 0.83) and obstructive codes (J44/J45/J46 combined: 43.1% vs. 67.0%; RR 0.64). Conclusions: In this retrospective analysis, an EMS unit with systematically implemented ultrasound demonstrated a lower frequency of hospital transport for selected dyspnea/respiratory failure codes compared with regional data and greater precision in ICD-10 diagnostic coding in cases where ultrasound was performed. The profile of LUS findings correlated with clinical categories in a manner consistent with existing literature. Full article

Background: Photodynamic therapy (PDT) has emerged as a powerful minimally invasive modality for cancer treatment. However, its efficacy as a monotherapy is often limited by oxygen dependence and limited light penetration. Combining PDT with systemic anticancer drug therapies offers a promising strategy to achieve synergistic effects and overcome resistance. Objective: This review aims to provide a systematic analysis of the mechanisms and clinical potential of combining PDT with chemotherapy, targeted therapy, and immunotherapy, focusing on recent advancements and nanotechnology-based delivery systems. Methods: A comprehensive literature search was performed using PubMed and Scopus databases. The analysis focused on peer-reviewed studies published over the last 10 years addressing synergistic molecular pathways, co-delivery nanoplatforms, and clinical trial outcomes. Results: The combination of PDT with chemotherapy enhances drug accumulation via vascular photosensitization and can overcome multi-drug resistance. Integration with immunotherapy, particularly immune checkpoint inhibitors and tumor vaccines, triggers immunogenic cell death (ICD), leading to systemic antitumor responses. Nanotechnology provides a versatile platform for the targeted co-delivery of photosensitizers and pharmacological agents, significantly reducing systemic toxicity. Conclusions: Combined PDT–drug regimens demonstrate superior therapeutic efficacy compared to monotherapies. Future clinical translation requires the standardization of dosimetry and the development of multifunctional nanomedicines to enable personalized treatment protocols. Full article

Background: Cemented and cementless fixation techniques in total hip arthroplasty (THA) each present distinct biomechanical properties and perioperative risk profiles. While cementless fixation has gained increasing popularity, large-scale nationally representative comparisons of perioperative outcomes between cemented and cementless elective THA remain limited. This study aimed to compare complication rates, healthcare utilization, and temporal trends between cemented and cementless elective THA using the National Inpatient Sample. Methods: A retrospective cohort study was conducted using the National Inpatient Sample database from 2016 to 2021. Adult patients undergoing elective primary total hip arthroplasty were identified using ICD-10-PCS codes and categorized into cemented and cementless fixation groups. Patient demographics, comorbidities, indications, postoperative complications, length of stay, hospital charges, and in-hospital mortality were compared. Multivariate logistic regression analysis was performed to evaluate the independent association between fixation type and postoperative complications while adjusting for demographic, clinical, and hospital-level variables. Results: A total of 81,668 elective THAs were identified, including 40,290 cemented (49.33%) and 41,378 cementless (50.67%) procedures. Cemented THA was associated with a shorter length of stay (2.09 ± 1.88 vs. 2.26 ± 2.47 days, p < 0.001) and lower total hospital charges ($65,584.53 ± 48,797.21 vs. $72,186.84 ± 49,860.20, p < 0.001). Unadjusted analyses demonstrated higher rates of acute kidney injury and sepsis in the cementless group. After multivariate adjustment, cemented fixation was associated with lower odds of acute kidney injury (OR 0.87, 95% CI 0.79–0.96, p = 0.004). However, cemented THA was associated with higher odds of postoperative delirium (OR 1.20, 95% CI 1.02–1.42, p = 0.030), blood transfusion (OR 1.27, 95% CI 1.17–1.37, p < 0.001), and periprosthetic fracture (OR 1.32, 95% CI 1.02–1.71, p = 0.035). Rates of myocardial infarction, pneumonia, venous thromboembolism, urinary tract infection, and in-hospital mortality were similar between groups. Temporal analysis demonstrated comparable utilization trends, with a decline in elective procedures during 2020–2021. Conclusions: In this nationwide analysis, cemented total hip arthroplasty was associated with lower risk of acute kidney injury, shorter length of stay, and lower hospital charges, but higher odds of postoperative delirium, blood transfusion, and periprosthetic fracture compared with cementless fixation. These findings highlight distinct perioperative risk profiles between fixation strategies and may assist surgeons in individualized decision-making for elective total hip arthroplasty. Full article

Background: Functional somatic symptoms (FSSs) represent a significant clinical challenge in pediatric populations, with prevalence estimates ranging from 10 to 30%. The COVID-19 pandemic introduced unprecedented psychosocial stressors that may have influenced the presentation and precipitating factors of these conditions. This article aimed to characterize the clinical presentations, precipitating factors, and temporal patterns of functional somatic disorders in children admitted to a tertiary pediatric hospital in Bucharest, Romania, across pre-pandemic (2017–2019), pandemic (2020–2022), and post-pandemic (2023–2025) periods. Methods: This retrospective study included 1043 patients aged 3–17 years admitted with somatic symptoms without identifiable organic pathology. Data were extracted using ICD-10 diagnostic codes and confirmed through individual chart review. Variables analyzed included demographics, symptoms, precipitating factors, symptom duration, and family psychiatric history. Results: Female patients predominated (67.0%), with a median age of 14 years. Cardiovascular symptoms were most frequent (43.9%), followed by neurological (30.4%), digestive (19.0%), and respiratory (6.3%) manifestations. Family-related factors (38.6%) and school-related stress (32.7%) were the primary precipitating factors. Significant pandemic-related differences emerged: medical-related precipitating factors increased during the pandemic (18.5% vs. 10.7% pre-pandemic, p < 0.001), respiratory symptoms were more frequent (9.8% vs. 5.9% pre-pandemic), and symptom duration before admission was significantly longer (median 1 vs. 1 pre-pandemic and 0 months post-pandemic; p < 0.001). Conclusions: The COVID-19 pandemic substantially influenced pediatric somatic presentations, with increased health anxiety, respiratory symptoms, and delayed healthcare-seeking behavior. Post-pandemic patterns suggest the persistent influence of traditional stressors alongside pandemic-related effects. Full article

Background and Objectives: Cardiovascular disorders contribute substantially to medical ineligibility for compulsory military service in Lithuania. This study aimed to describe cardiovascular disease patterns and assess their association with military service eligibility among conscription-age individuals treated at a tertiary care center, considering gender, place of residence, age, and the most common cardiovascular causes of medical ineligibility. Materials and Methods: This retrospective hospital-based study included men and women aged 18–26 years with cardiovascular disease diagnoses defined according to ICD-10 codes specified by the Order of the Minister of National Defense of the Republic of Lithuania. Anonymized medical records from the Hospital of the Lithuanian University of Health Sciences Kaunas Clinics were reviewed. Participants were categorized into four military service eligibility classes based on medical history data and were stratified by gender, age, and place of residence. Results: The study included 521 participants (56.6% male, 43.4% female). Gender and residence showed no significant impact on military service eligibility. Younger individuals, particularly those aged 18–19, were more often deemed eligible, while eligibility declined with age. Males more commonly had essential hypertension and hypertensive heart disease, whereas females more frequently presented with paroxysmal tachycardia and other arrhythmias. Hypertension and other severe cardiovascular conditions most strongly reduced eligibility for compulsory military service, whereas rhythm disorders were more often compatible with service. Conclusions: In this hospital-based cohort of conscription-age individuals with cardiovascular disease, gender and place of residence did not significantly influence eligibility for military service. Eligibility declined with increasing age, and hypertension-related cardiovascular disorders were the leading cause of ineligibility among conscripts. Full article

In intensive care units, large-scale clinical time-series data are continuously accumulated through electronic medical records and bedside monitoring systems. However, direct utilization of such data for clinical decision-making remains challenging due to irregular sampling, pervasive missingness, unstructured diagnostic information, and incomplete ICD labeling. Automated ICD coding constitutes an extreme multi-class classification problem with thousands of long-tailed categories, while intervention prediction tasks, such as mechanical ventilation management, involve rare transition events and severe class imbalance. To address these challenges, we propose CAGE, a hierarchical Clinical Decision Support System framework that integrates diagnosis, time-series signals, and intervention prediction. The framework first infers admission-level diagnostic context using a partial-label Automated ICD Coding module that combines DCNv2 with an Adaptive CLPL loss, producing probability-weighted diagnostic embeddings. These embeddings are subsequently fused with ICU time-series tensors and processed by a multi-branch Temporal Convolutional Network equipped with an ICD-conditioned gating mechanism to predict future ventilation state transitions. The experimental results demonstrate that DCNv2 achieves consistent superiority across all hit@k and probability concentration metrics for ICD coding. For intervention prediction, the proposed method substantially outperforms existing baselines, achieving a Macro-AUC of 98.2, Macro-AUPRC of 77.4, and F1-score of 79.4. These findings indicate that reinjecting diagnostic context as a conditioning variable, together with imbalance-aware loss design, effectively enhances rare-event detection and improves the practical applicability of clinical decision support systems. Full article