Search Results (1,896)

Background/Objectives: Trastuzumab emtansine (T-DM1) is widely used in Human Epidermal Growth Factor Receptor2 (HER2)-positive metastatic breast cancer; however, outcomes vary, and reliable prognostic markers remain limited. We developed the CALA index as a composite biomarker integrating CA15-3, lactate dehydrogenase (LDH), and albumin. This study aimed to evaluate the prognostic value of the CALA index in metastatic breast cancer treated with T-DM1. Methods: This multicenter retrospective study included 168 patients treated with T-DM1 across four tertiary centers. The CALA index was calculated using pretreatment levels of CA15-3, LDH, and albumin. Receiver operating characteristic (ROC) curve analysis was used to determine the optimal cutoff value, and patients were stratified into groups accordingly. Survival outcomes and independent risk factors were assessed using Kaplan–Meier and Cox regression analyses. Results: The median overall survival (OS) was 26 months (95% CI: 21.3–30.7). ROC analysis identified an optimal CALA cutoff value of 118.3. Patients with CALA ≤ 118.3 demonstrated significantly longer OS compared with those with CALA > 118.3 (log-rank p = 0.006), with 1- and 3-year OS rates of 81.2% and 43.2% versus 69.8% and 22.7%, respectively. In univariate analysis, CALA > 118.3 was associated with worse OS (HR: 1.699; 95% CI: 1.151–2.506; p = 0.008), and this association remained significant in multivariate analysis (HR: 1.671; 95% CI: 1.088–2.565; p = 0.019). Conclusions: The CALA index was associated with overall survival in metastatic breast cancer treated with trastuzumab emtansine and may serve as a practical tool for risk stratification. Full article

Background/Objectives: Early identification of ST-segment elevation myocardial infarction (STEMI) at first medical contact remains challenging, as high-sensitivity troponin T may be insufficiently sensitive during the initial phase of myocardial injury. Readily available complete blood count (CBC)-derived inflammatory indices may provide complementary early diagnostic signals. This study aimed to evaluate whether baseline CBC-derived inflammatory indices differ between STEMI and NSTEMI and whether they provide adjunctive discriminatory information at presentation (0 h) in patients with acute coronary syndrome (ACS). Methods: A 12-lead electrocardiogram (ECG), high-sensitivity troponin T, and CBC were obtained at presentation from 252 patients with ACS (195 STEMI and 57 NSTEMI). Diagnostic performance was evaluated using receiver operating characteristic (ROC) curve analysis and 2 × 2 contingency tables to determine the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and likelihood ratios. Results: High-sensitivity troponin T demonstrated the highest specificity (84.44%) and PPV (92.93%), supporting its role as a confirmatory biomarker; however, its low sensitivity (50.83%) and NPV (29.92%) may reduce its utility during early assessment. In contrast, WBC and neutrophil counts demonstrated relatively favorable discriminatory performance at presentation (AUC > 0.72; Youden’s index > 0.40). Among composite indices, NLPR demonstrated the highest sensitivity (88.66%) and NPV (53.19%), along with the lowest negative likelihood ratio (0.25), suggesting potential adjunctive value during early assessment. NLR, SII, SIRI, and adjusted NLR showed moderate performance, with aNLR providing a balanced sensitivity (67.01%) and specificity (74.55%). Conclusions: CBC-derived inflammatory indices, particularly neutrophil-based markers such as NLPR, may provide adjunctive discriminatory information during the early assessment of patients with ACS, particularly at first medical contact when baseline hs-Troponin T sensitivity may still be limited. Full article

Background/Objectives: Preventing recurrent anaphylaxis is crucial for patient safety. This study aimed to identify predictive factors and develop a prediction model to estimate recurrence risk, thereby enhancing targeted preventive strategies. Methods: This prognostic prediction study used a retrospective observational cohort design, analyzing medical records from an anaphylaxis registry at Naresuan University Hospital, Phitsanulok, Thailand, between March 2011 and February 2021. We developed a prediction model using multivariable Cox proportional hazards regression analysis. Statistically significant and clinically relevant predictors were weighted into a risk score derived from hazard ratio regression coefficients. Model performance was evaluated using the area under the receiver operating characteristic curve (AuROC), calibration metrics, and decision curve analysis. Results: Over the 10-year period, 381 patients experienced 439 anaphylaxis episodes, including 58 recurrences (13.2%). The final model comprised six predictors: history of food, insect, and drug allergies; asthma; chest discomfort; and severe anaphylaxis. Corresponding risk scores were 4, 5, 5.5, 1, 2.5, and 1.5 points, respectively. Total scores ranged from 0 to 19.5 and were categorized into low (<3.0), moderate (3.0–9.0), and high (>9.0) risk groups. The high-risk group had a likelihood ratio positive (LHR+) of 4.65. The model demonstrated acceptable discrimination (AuROC 0.773 (95% CI: 0.714–0.832)) and good calibration. Bootstrap validation showed consistent performance (AuROC 0.773 (95% CI: 0.714–0.831)). Decision curve analysis indicated clinical utility across relevant threshold probabilities. Conclusions: This prediction model provides a simple, clinically applicable tool for estimating the risk of recurrent anaphylaxis and may support improved prevention and management strategies. Full article

Background: The immune mechanism of diabetic kidney disease (DKD) has not yet been fully elucidated. This study aimed to characterize circulating lymphocyte subsets in patients with type 2 diabetes mellitus (T2DM), with a particular focus on DKD-related immune alterations and prognosis. Methods: Circulating T cells, B cells and NK cells were identified by flow cytometry. The primary endpoint was all-cause mortality, and overall survival was defined as the time from enrollment to death from any cause or last follow-up. Associations between lymphocyte subsets, inflammatory indices and renal function parameters were analyzed. Cox regression was used to identify factors associated with overall survival in patients with DKD and in the whole T2DM cohort. A prognostic nomogram was developed in the whole T2DM cohort to estimate 1-, 2-, 3-, and 5-year overall survival (OS) probabilities. Model performance was evaluated using the concordance index (C-index), calibration curves, receiver operating characteristic (ROC) curves, and decision curve analysis (DCA). Mendelian randomization (MR) was performed as a further exploratory analysis to assess whether immune-related traits were genetically associated with DKD susceptibility, with inverse variance weighting (IVW) as the primary analytical method. Results: In total, 74 T2DM patients were divided into DKD (stage 3–4 of chronic kidney disease) and non-DKD groups. Median follow-up duration was 34.6 months. DKD patients exhibited elevated levels of NK cells, the monocyte-to-lymphocyte ratio (MLR), neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR). In patients with DKD, higher PLR and serum creatinine (SCr) were associated with poorer overall survival, whereas CD4⁺CD25⁺ T cell frequency was not significant after adjustment. In the whole T2DM cohort, higher frequency of circulating CD4⁺CD25⁺ T cells were associated with improved survival (HR 0.920, 95% CI 0.858–0.986, p = 0.019), whereas elevated PLR and SCr were linked to poorer outcomes. The exploratory nomogram incorporating CD4⁺CD25⁺ T cells, PLR, and SCr, showed acceptable internal performance in this cohort. As a separate exploratory analysis, MR suggested that genetically proxied CD4 expression on activated CD4 regulatory T cells was associated with a lower risk of DKD. Conclusions: DKD was associated with higher mortality and elevated MLR-, NLR-, PLR-, and NK cell levels in patients with T2DM. In patients with DKD, PLR and SCr were associated with overall survival, supporting the prognostic relevance of systemic inflammation and renal dysfunction. Individual lymphocyte subsets were not independently associated with survival in the DKD cohort after adjustment, whereas CD4⁺CD25⁺ T cell frequency provided additional prognostic information in the whole extended T2DM cohort analysis. Further validation is warranted. Full article

Background and Objectives: Obstructive sleep apnea (OSA) is a common disorder associated with significant cardiovascular, metabolic, and neurocognitive consequences. The STOP and STOP-BANG questionnaires are widely used screening tools for identifying individuals at increased risk of OSA. However, their performance may vary across populations. This variability is due to demographic and anthropometric differences. We aimed to analyze the screening accuracy of the STOP and STOP-BANG questionnaires across three distinct Mediterranean populations: Croatia, Greece, and Türkiye. Additionally, we aimed to optimize and establish population-specific cut-off points for body mass index (BMI) and neck circumference (NC) in the questionnaires to enhance their screening accuracy. Materials and Methods: A total of 9102 patients who underwent polysomnography or polygraphy to evaluate suspected OSA were enrolled from: Split Sleep Medicine Centre (Croatia), Ege University Faculty of Medicine (Türkiye), and Thessaloniki G Papanikolaou Hospital Aristotle University (Greece). Patients completed the STOP and STOP-BANG questionnaires before sleep assessments. Sensitivity, specificity, and the area under the receiver operating characteristic (ROC) curve (AUC) were calculated to assess the screening properties. Additionally, optimized cut-offs for age, NC, and BMI were determined. Results: The highest AUC values were observed using the STOP-BANG ≥ 5 method, with AUC values of 0.712 for detecting any OSA (AHI ≥ 5/h), 0.684 for moderate or severe OSA (AHI ≥ 15/h), and 0.663 for severe OSA (AHI ≥ 30/h). For individual centers, the STOP-BANG ≥ 5 method performed best in Split, while the STOP ≥ 2 + NC method yielded the highest AUCs in Izmir and Thessaloniki for moderate and severe OSA. Optimized cut-off values for age, NC, and BMI improved sensitivity and specificity across all centers. Conclusions: This study highlights the need for population-specific considerations in the screening for OSA. Significant differences in demographics, anthropometrics, symptoms, and comorbidities across populations could impact the questionnaire’s screening accuracy. Adjusting age, NC, and BMI cut-off points optimizes the STOP-BANG questionnaire. Full article

Objective: Isolated gestational proteinuria (IGP) has traditionally been considered a benign condition; however, emerging evidence suggests that it may represent an early stage in the spectrum of preeclampsia. This study aimed to evaluate clinical and laboratory predictors of progression from IGP to preeclampsia. Methods: This retrospective cohort study included pregnant women diagnosed with proteinuria ≥ 300 mg/day after 20 weeks of gestation between January 2023 and December 2024. After applying predefined exclusion criteria, 319 women with isolated gestational proteinuria (IGP) were included and stratified according to progression to preeclampsia (n = 42, 17.8%). Baseline clinical and laboratory parameters were compared between groups. Multivariable logistic regression analysis was performed to identify independent predictors of progression, and receiver operating characteristic (ROC) curve analysis was used to evaluate the discriminative performance of significant variables. Results: Preeclampsia developed in 17.8% of women with IGP. In multivariable analysis, higher maternal body mass index (aOR 1.085, p = 0.028) and earlier gestational age at diagnosis (aOR 0.883, p = 0.011) were identified as independent predictors of progression. Although neutrophil count and systemic inflammatory indices were elevated in univariate analyses, they did not retain independent predictive value after adjustment. Conclusions: In pregnancies complicated by isolated gestational proteinuria, clinical parameters appear to be more informative than inflammatory and renal markers for predicting progression to preeclampsia. Laboratory-derived indices offer limited additional value and should be interpreted cautiously in risk assessment. Full article

Background/Objectives: Vaginal cuff hematoma is a recognized complication following hysterectomy, with a subset of patients requiring invasive intervention. No reliable bedside biomarker currently exists to identify at admission patients likely to fail conservative management. This study aimed to evaluate the incidence and clinical characteristics of symptomatic vaginal cuff hematoma across all hysterectomy approaches, and to assess the predictive performance of the CALLY index (CRP-albumin-lymphocyte index), a composite marker of inflammatory burden, immune function, and nutritional status, alongside the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and systemic immune-inflammation index (SII) for identifying patients requiring interventional management. Methods: This retrospective cohort study included 61 patients with symptomatic vaginal cuff hematoma following hysterectomy in a major tertiary referral center (November 2022–July 2025). Patients were divided into conservative (n = 38) and interventional (n = 23) management groups. The CALLY index was calculated as [Albumin (g/dL) × Lymphocyte (×10⁹/L)] ÷ [CRP (mg/L) × 10⁻²]. Receiver operating characteristic (ROC) curve analysis with the DeLong method was used to compare predictive performance. Results: The overall incidence of symptomatic vaginal cuff hematoma was 1.9% (73/3852 hysterectomies), with the highest rate following vaginal hysterectomy (3.32%) and the lowest after robotic hysterectomy (0.74%). Interventional management was required in 37.7% of patients. The interventional group had significantly higher CRP (192 vs. 62 mg/L, p < 0.001), NLR (7.53 vs. 4.17, p < 0.001), and SII (2308 vs. 1207, p < 0.001), and significantly lower CALLY index values (2.00 vs. 9.80, p < 0.001). The CALLY index demonstrated the highest predictive performance (AUC = 0.863, 95% CI: 0.762–0.964), outperforming SII (AUC = 0.801), NLR (AUC = 0.789), and PLR (AUC = 0.654). At the optimal cutoff of ≤2.89, the CALLY index yielded a sensitivity of 65.2% and a specificity of 92.1%. Conclusions: The CALLY index is a simple, routinely available composite biomarker that may help identify patients at higher risk for interventional management in symptomatic vaginal cuff hematoma. Its incorporation into postoperative assessment may improve risk stratification and support timely clinical decision-making. Full article

Background and Objectives: This study evaluated the effect of two wound closure methods—polypropylene sutures and a butyl-2-cyanoacrylate tissue adhesive—on the rate of soft tissue regeneration following palatal donor site harvesting. A bovine collagen sponge, used as a secondary-intention dressing, was evaluated descriptively. Materials and Methods: Data from 300 patients (n = 100/group) with palatal donor sites were analyzed. Primary analysis compared suture vs. adhesive using Early Wound Healing Score (EHS) at days 7 and 14. Secondary outcomes were granulation tissue (day 7) and complications. Statistical methods: Mann–Whitney U test for between-group comparison (suture vs. adhesive); Kruskal–Wallis with Dunn’s post hoc for granulation across all three groups; Spearman’s correlation and logistic regression for the relationship between granulation tissue and EHS within primary healing groups. Results: At day 7, median EHS was similar between suture and adhesive groups (7.0 [interquartile range (IQR) 5.0–9.0] vs. 7.0 [IQR 7.0–9.0]; p = 0.31). By day 14, both groups achieved excellent healing (median 10.0, IQR 9.0–10.0 in both; p = 0.82). The collagen sponge group showed slower healing (median EHS day 7 = 4.0 [IQR 3.0–5.0], day 14 = 6.0 [IQR 5.0–7.0]), reported descriptively as expected for secondary intention. Granulation tissue on day 7 was highest in the adhesive group (p < 0.001 vs. collagen; p = 0.024 vs. suture). A strong positive correlation between day-7 granulation tissue and day-14 EHS was found in the primary-healing groups (ρ = 0.78, p < 0.001). Receiver operating characteristic (ROC) analysis established a granulation score ≥ 2 as the optimal cut-off for predicting successful healing (EHS ≥ 9) by day 14 (sensitivity 89.4%, specificity 76.0%, area under the curve (AUC) = 0.80), pending external validation. Conclusions: Surgical adhesive may be considered a viable alternative to sutures for palatal donor sites closed by primary intention, offering comparable healing by day 14. Collagen sponges result in slower healing and should be considered only when secondary intention is specifically desired. Early assessment of granulation tissue may serve as a simple prognostic indicator, but external validation is needed before clinical application. Full article

Background/Objectives: Urinary calcium excretion is a key parameter in assessing mineral metabolism and diagnosing conditions such as nephrolithiasis, osteoporosis, and hyperparathyroidism. The 24 h urine collection is the gold standard for evaluating calcium excretion, but it is often impractical due to patient non-compliance and logistical challenges. As an alternative, the calcium-to-creatinine ratio (CCR) in spot urine has been proposed, although its reliability remains debated. This study aims to systematically compare the calcium levels in spot urine samples with those obtained from 24 h urine collections to assess their agreement and clinical applicability. Methods: This retrospective, multi-center study analyzed data from 201 patients who provided both 24 h and spot urine samples during routine diagnostic work-up between 1 January 2019 and 31 December 2024. Calcium excretion was normalized using the calcium-to-creatinine ratio (CCR). The agreement between the two methods was assessed using Bland–Altman analysis, Pearson and Spearman correlation coefficients, and receiver operating characteristic (ROC) curve analysis. Results: Hypercalciuria, defined as ≥6.25 mmol/24 h in women and ≥7.5 mmol/24 h in men, was detected in 52.7% of cases based on 24 h urine. ROC analysis showed that spot urine CCR had moderate diagnostic accuracy (AUC = 0.76). The optimal cut-off for predicting hypercalciuria was 4.4 mmol/g (sensitivity 70.8%, specificity 72.4%). Overall agreement between spot urine CCR and 24 h urine CCR was moderate, with a Bland–Altman geometric mean ratio of 1.06 and multiplicative limits of agreement of 0.59 to 1.91. A low spot urine CCR below 2 mmol/g showed high sensitivity but low specificity and had a negative predictive value of 82%. Conclusions: Spot urine CCR cannot replace 24 h urine collection for accurately assessing urinary calcium excretion, but very low values may have limited utility as an initial rule-out tool in selected patients. Very low spot urine CCR values may help rule out hypercalciuria in a limited subgroup of patients and may therefore support triage decisions in selected clinical situations. Further prospective studies are needed to validate these findings. Full article

One of the major determinants of crop productivity and sustainable agricultural practices is soil fertility. Proper soil assessment helps farmers make informed decisions about nutrients and fertilizers. This study utilizes 16 machine learning classifiers for soil fertility prediction, including learner-based, ensemble-based, instance-based, and probabilistic-based models. The model’s performance is assessed using accuracy, precision, recall, and F1-score. This paper presents a machine learning model for predicting soil fertility based on soil physicochemical characteristics. The data used in the research comprise vital soil parameters: nitrogen, phosphorus, potassium, pH, organic carbon, electrical conductivity, and micronutrients. Missing-value imputation, label encoding, and feature standardization are among the data preprocessing methods used to enhance data quality. Correlation analysis, ANOVA F-score, and mutual information were used to assess feature importance and determine the most significant soil characteristics. The experimental observation reveals that the RF model achieves an accuracy of 90.91% compared to the other models. Additional assessment using multi-class Receiver Operating Characteristic (ROC) and Precision–Recall (PR) curves showed excellent discriminative ability across the dominant soil fertility, which was of high quality. The findings show that machine learning models, especially ensemble-based models, are effective at estimating soil fertility levels. The proposed framework provides a data-driven, reliable decision-support system to assess soil fertility, enabling farmers and agricultural experts to enhance nutrient management and crop production. Full article

Objective: To explore the feasibility and effectiveness of an enhanced CT deep learning model based on regional attention for the preoperative multi-classification of rectal cancer T stages. Methods: Five hundred eligible patients with rectal cancer (48 in T1 stage, 127 in T2, 259 in T3, and 66 in T4) were randomly divided into a training group (n = 400) and a validation group (n = 100). Regions of interest (ROIs) in rectal cancer lesions were pixel-wise annotated by experienced radiologists. A deep learning algorithm based on regional attention was used to train a binary classification model (early stage—T1 and T2, advanced stage—T3 and T4) and a multi-classification model (T1, T2, T3 and T4 stages), which were compared against radiomics approaches. Features were extracted from manually segmented ROIs using pyradiomics, radiomics-based binary and multi-classification models using ten different algorithms. In addition, baseline clinical data-based binary and multi-classification models were also constructed. The performance of both binary and multi-classification models were evaluated by plotting receiver operating characteristic (ROC) curves. The area under the curve (AUC) and accuracy were calculated for the binary model, and the micro-average AUC, macro-average AUC, and accuracy were calculated for the multi-classification model. Results: The ROI-based binary classification model for T stage (ROITransStage; AUC = 0.878, accuracy = 0.850) outperformed the best among ten radiomics-based binary models (AdaBoost; AUC = 0.802, accuracy = 0.76), as well as the best-performing baseline clinical data binary model (AdaBoost; AUC = 0.836, accuracy = 0.76). In addition, ROITransStage (micro-average AUC = 0.873, macro-average AUC = 0.862, accuracy = 0.81) also demonstrated superior diagnostic performance for the T1, T2, T3 and T4 stages compared to the best-performing radiomics-based (SVM; micro-average AUC = 0.845, macro-average AUC = 0.777, accuracy = 0.6) and baseline clinical data-based (SVM; micro-average AUC = 0.841, macro-average AUC = 0.76, accuracy = 0.61) multi-classification models. Conclusions: The CT deep learning binary and multi-classification models based on regional attention exhibited superior predictive performance for rectal cancer staging compared to both radiomics and clinical data-based models. Full article

Background: Interventional occlusion procedures for congenital heart disease (CHD) carry the risk of complications requiring reintervention, yet predictive factors remain unclear. Methods: This retrospective case–control study included patients (n = 4190) with simple CHD who underwent transcatheter/interventional procedure (2017–2022). Perioperative and postoperative complications were monitored at 1, 3, and 6 months after occlusion. Among them, 44 patients required reintervention for complications. Statistical analysis was performed on clinical data, ultrasound findings from various locations, and laboratory examination results. Results: For atrial septal defects (ASD), independent predictors were defect size and age grading, while those for ventricular septal defects (VSD) were occluder device size, aortic annulus inner diameter, body surface area class, and whether the defect was isolated. The areas under the curve (AUC) of the receiver operating characteristic (ROC) curve for patients who experienced severe complications requiring surgical repair according to ASD were 0.723, whereas for VSD, the AUCs for occluder device size and aortic valve annulus diameter among patients who experienced severe complications requiring surgical repair were 0.649 and 0.539, respectively. Conclusions: This study provides an inaugural comprehensive analysis of occurrence rates and predictive factors for severe post-interventional occlusion procedure complications requiring reintervention. These findings offer new insights as a reference for the treatment of CHD. Full article

Background/Objectives: Machine learning approaches are widely used in modern medical diagnostics, including cancer detection. The results can be significantly improved by aggregating individual measurements, and appropriate aggregation methods should be established. Methods: We applied various measurement aggregation strategies both before and after machine learning modeling to two datasets of X-ray diffraction images: human breast biopsy samples and canine claw samples. Two classifiers, Random Forest and Logistic Regression, were used to determine classification metrics: the area under the receiver operating characteristic curve (ROC-AUC) and balanced accuracy. Results: We found that all aggregation types improve classification metrics, with aggregation after modeling yielding better performance. Depending on the dataset and approach, either classifier can produce better results. For human breast samples, Random Forest with the logit aggregation strategy provides an ROC-AUC exceeding 0.9. For the canine dataset, both Random Forest with the logit aggregation strategy and Logistic Regression with the median of cancer probabilities achieve an ROC-AUC of about 0.85. Conclusions: We examined several simple, straightforward aggregation methods for patient diagnosis based on multiple measurements per patient and achieved significant improvements in classification metrics. Full article

Background/Objectives: This study assessed the diagnostic accuracy of two commonly used diagnostic instruments for autism spectrum disorder (ASD), the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2) and the Autism Diagnostic Interview-Revised (ADI-R), in comparison to a best-estimate (BE) diagnosis made by a research psychologist. Methods: Two hundred and thirteen children aged 5 years 0 months to 7 years 11 months completed a comprehensive research assessment that included multiple diagnostic measures. Once each research assessment was complete, a research psychologist gave each participant an overall BE research diagnosis of Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) ASD based on all available information from diagnostic testing and behavioral observations during testing. We assessed sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of both the ADOS-2 and ADI-R separately and in combination and used receiver operating characteristic (ROC) curves to compare the areas under the curve (AUCs) of these instruments. Results: Both the ADOS-2 Spectrum Criterion scoring (sensitivity = 96.2%; specificity = 97.5%) and ADOS-2 Autism Criterion scoring (sensitivity = 82.0%; specificity = 100%) had excellent accuracy in comparison to the BE ASD diagnosis. The ADI-R had good accuracy (sensitivity = 78.6%; specificity = 83.5%) compared to BE ASD diagnosis. In receiver operating curve analyses, both scoring criteria for ADOS-2 were significantly more accurate than the ADI-R. Conclusions: Overall, both instruments provide good, if not excellent, classification accuracies when used individually, as well as in combination. Thus, when deciding which measures to use for ASD research, other factors should also be considered. Full article

Background/Objectives: The aim of this pilot study was to evaluate the feasibility of developing individualized machine learning models using nocturnal wearable-derived autonomic nervous system (ANS) and sleep metrics to predict next-day headache risk in patients with migraine. We also examined the associations between nocturnal ANS and sleep measures and patient-reported outcome measures (PROMs) related to nociplastic pain, migraine burden, and non-restorative sleep (NRS). Methods: Adults with migraine wore the wrist-worn Empatica EmbracePlus^® wearable during sleep and completed daily headache diaries for approximately 4 weeks (N = 10). Participants also completed daily headache diaries and PROMs assessing nociplastic pain, migraine burden, and non-restorative sleep. Personalized machine learning (ML) models were developed to predict next-day headache using nocturnal ANS activity (e.g., pulse rate variability (PRV), electrodermal activity (EDA), respiratory rate (RR)) and sleep metrics (e.g., interruptions, duration, awakenings). Model performance was evaluated using area under the receiver operating characteristic and precision–recall curves (AUROC, AUPRC), sensitivity, specificity, accuracy, and precision. Spearman correlations assessed the relationship between wearable-derived metrics and patient-reported outcome measurements of sleep quality (PROMIS-Fatigue, PROMIS-Sleep Disturbance) and a surrogate marker of nociplastic pain (Fibromyalgia (FM) Score). Results: 9 out of 10 participants wore the EmbracePlus device for at least the target duration of four weeks. For the next-day headache prediction, model performance varied between individuals; area under the ROC curve (AUROC) ranged from 28.2% to 81.2%. Nocturnal measures of EDA were strongly correlated with the FM score (Spearman’s rho = 0.72–0.75, p < 0.05). Conclusions: Phasic EDA may warrant further investigation as a potential physiological indicator related to nociplastic pain mechanisms and next-day headache. However, these findings are preliminary, and larger multicenter trials are needed to confirm results of this pilot study. Full article