Background: Since cancer pain requires complex modalities of care, the proper strategy for addressing its telemedicine-based management should be better defined. This study aimed to trace a pathway for a progressive implementation of the telemedicine process for the treatment of pain in the setting of cancer patients. Methods: The features of the model were investigated to dissect the dropout from the telemedicine pathway. A cross-sectional patient satisfaction study was conducted. The degree of satisfaction was evaluated through a developed 22-item questionnaire (Likert scale 0–7). Results: A total of 375 video consultations for 164 patients (mean age 62.9 years) were performed through remote consultations for cancer pain management between March 2021 and February 2022. After the exclusion of 72 patients, 92 (56.1%) were included in the analysis. The dropout ratio was 8.7%. The number of visits and pharmacological therapies for neuropathic pain correlated with the risk for readmission (p < 0.05). Overall, the satisfaction was very high (mean > 5.5 for all items). Conclusion: Feedback from patients reflected high satisfaction rates with the care provided. A methodological approach based on the degree of satisfaction combined with the analysis of the pathways can help to implement the quality of a service provided through telemedicine. While not without limitations, our hybrid protocol can be useful for addressing cancer pain through a patient-centered approach.

Background: The most effective strategy for managing cancer pain remotely should be better defined. There is a need to identify those patients who require increased attention and calibrated follow-up programs. Methods: Machine learning (ML) models were developed using the data prospectively obtained from a single-center program of telemedicine-based cancer pain management. These models included random forest (RF), gradient boosting machine (GBM), artificial neural network (ANN), and the LASSO–RIDGE algorithm. Thirteen demographic, social, clinical, and therapeutic variables were adopted to define the conditions that can affect the number of teleconsultations. After ML validation, the risk analysis for more than one remote consultation was assessed in target individuals. Results: The data from 158 patients were collected. In the training set, the accuracy was about 95% and 98% for ANN and RF, respectively. Nevertheless, the best accuracy on the test set was obtained with RF (70%). The ML-based simulations showed that young age (75 years) with bone metastases may require more telemedicine-based clinical evaluations. Conclusion: ML-based analyses may enable clinicians to identify the best model for predicting the need for more remote consultations. It could be useful for calibrating care interventions and resource allocation.

Background/Objectives: Cancer and related treatments can impair body composition (BC), increasing the risk of malnutrition and sarcopenia, poor prognosis, and Health-Related Quality of Life (HRQoL). To enhance BC parameter interpretation through Bioelectrical Impedance Analysis (BIA), we developed a predictive model based on unsupervised approaches including Principal Component Analysis (PCA) and k-means clustering for sarcopenia risk in cancer patients at the Istituto Nazionale Tumori IRCCS “Fondazione G. Pascale” (Naples). Methods: Sarcopenia and malnutrition risks were assessed using the NRS-2002 and SARC-F questionnaires, anthropometric measurements, and BIA. HRQoL was evaluated with the EORTC QLQ-C30 questionnaire. PCA and clustering analysis were performed to identify different BC profiles. Results: Data from 879 cancer patients (mean age: 63 ± 12.5 years) were collected: 117 patients (13%) and 128 (15%) were at risk of malnutrition and sarcopenia, respectively. PCA analysis identified three main components, and k-means determined three clusters, namely HMP (High Muscle Profile), MMP (Moderate Muscle Profile), and LMP (Low Muscle Profile). Patients in LMP were older, with a higher prevalence of comorbidities, malnutrition, and sarcopenia. In the multivariable analysis, age, lung cancer site, diabetes, and malnutrition risk were significantly associated with an increased risk of sarcopenia; among the clusters, patients in LMP had an increased risk of sarcopenia (+62%, p = 0.006). Conclusions: The NUTRISCREEN project, part of the ONCOCAMP study (ClinicalTrials.gov ID: NCT06270602), provides a personalized nutritional pathway for early screening of malnutrition and sarcopenia. Using an unsupervised approach, we provide distinct BC profiles and valuable insights into the factors associated with sarcopenia risk. This approach in clinical practice could help define risk categories, ensure the most appropriate nutritional strategies, and improve patient outcomes by providing data-driven care.

Background/Objectives: Deep learning (DL) has shown strong potential in analyzing food images, but few studies have directly predicted mass, energy, and macronutrient content from images. In addition to the importance of high-quality data, differences in country-specific food composition databases (FCDBs) can hinder model generalization. Methods: We assessed the performance of several standard DL models using four ground truth datasets derived from Nutrition5k—the largest image–nutrition dataset with ~5000 complex US cafeteria dishes. In light of developing an Italian dietary assessment tool, these datasets varied by FCDB alignment (Italian vs. US) and data curation (ingredient–mass correction and frame filtering on the test set). We evaluated combinations of four feature extractors [ResNet-50 (R50), ResNet-101 (R101), InceptionV3 (IncV3), and Vision Transformer-B-16 (ViT-B-16)] with two regression networks (2+1 and 2+2), using IncV3_2+2 as the benchmark. Descriptive statistics (percentages of agreement, unweighted Cohen’s kappa, and Bland–Altman plots) and standard regression metrics were used to compare predicted and ground truth nutritional composition. Dishes mispredicted by ≥7 algorithms were analyzed separately. Results: R50, R101, and ViT-B-16 consistently outperformed the benchmark across all datasets. Specifically, when replacing it with these top algorithms, reductions in median Mean Absolute Percentage Errors were 6.2% for mass, 6.4% for energy, 12.3% for fat, and 33.1% and 40.2% for protein and carbohydrates. Ingredient–mass correction substantially improved prediction metrics (6–42% when considering the top algorithms), while frame filtering had a more limited effect (<3%). Performance was consistently poor across most models for complex salads, chicken-based or eggs-based dishes, and Western-inspired breakfasts. Conclusions: The R101 and ViT-B-16 architectures will be prioritized in future analyses, where ingredient–mass correction and automated frame filtering methods will be considered.

Background: Unplanned hospital readmissions (HRAs) are very common in cancer patients. These events can potentially impair the patients’ health-related quality of life and increase cancer care costs. In this study, data-driven prediction models were developed for identifying patients at a higher risk for HRA. Methods: A large dataset on cancer pain and additional data from clinical registries were used for conducting a Bayesian network analysis. A cohort of gastrointestinal cancer patients was selected. Logical and clinical relationships were a priori established to define and associate the considered variables including cancer type, body mass index (BMI), bone metastasis, serum albumin, nutritional support, breakthrough cancer pain (BTcP), and radiotherapy. Results: The best model (Bayesian Information Criterion) demonstrated that, in the investigated setting, unplanned HRAs are directly related to nutritional support (p = 0.05) and radiotherapy. On the contrary, BTcP did not significantly affect HRAs. Nevertheless, the correlation between variables showed that when BMI ≥ 25 kg/m², the spontaneous BTcP is more predictive for HRAs. Conclusions: Whilst not without limitations, a Bayesian model, combined with a careful selection of clinical variables, can represent a valid strategy for predicting unexpected HRA events in cancer patients. These findings could be useful for calibrating care interventions and implementing processes of resource allocation.

Health-related quality of life (HRQoL) represents one of the most concerning aspects for cancer patients. The Healthy Eating Index (HEI) is an a priori diet quality index directly associated with health outcomes and HRQoL in cancer survivors in North American populations. We evaluated, in a Mediterranean population, the baseline associations between HEI-2015 and HRQoL in 492 women with breast cancer recruited in a DEDiCa lifestyle trial. Dietary data were obtained from 7-day food records; HRQoL was assessed through the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ C30) and the C30 Summary Score (SumSc). Analysis of variance and multivariable linear and log-gamma regression models were performed. Mean and standard deviation for HEI-2015 score was 68.8 ± 11.2; SumSc was 81.5 ± 12.9. Women with lower HEI-2015 score had higher BMI, were more frequently exposed to tobacco smoke and had fewer years of education. Patients with a HEI-2015 score greater than 68.7 (median value) showed a significant increase in SumSc of 4% (p = 0.02). HEI-2015 components also associated with SumSc were beans and greens (β = 1.04; p = 0.02). Weak associations were found for total vegetables and saturated fats. Higher diet quality in breast cancer survivors was associated with higher overall HRQoL in this cross-sectional analysis.