The Role of Anthocyanins in Alleviating Intestinal Diseases: A Mini Review.

Anthocyanins are phytonutrients with physiological activity belonging to the flavonoid family whose transport and absorption in the human body follow specific pathways. In the upper gastrointestinal tract, anthocyanins are rarely absorbed intact by active transporters, with most reaching the colon, where bacteria convert them into metabolites. There is mounting evidence that anthocyanins can be used for prevention and treatment of intestinal diseases, including inflammatory bowel disease (IBD), irritable bowel syndrome (IBS), and colorectal cancer (CRC), through the protective function on the intestinal epithelial barrier, immunomodulation, antioxidants, and gut microbiota metabolism. Dietary anthocyanins are summarized in this comprehensive review with respect to their classification and structure as well as their absorption and transport mechanisms within the gastrointestinal tract. Additionally, the review delves into the role and mechanism of anthocyanins in treating common intestinal diseases. These insights will deepen our understanding of the potential benefits of natural anthocyanins for intestinal disorders.

Optimized reconstruction of the absorption spectra of kidney tissues from the spectra of tissue components using the least squares method.

With the objective of developing new methods to acquire diagnostic information, the reconstruction of the broadband absorption coefficient spectra (µa [λ]) of healthy and chromophobe renal cell carcinoma kidney tissues was performed. By performing a weighted sum of the absorption spectra of proteins, DNA, oxygenated, and deoxygenated hemoglobin, lipids, water, melanin, and lipofuscin, it was possible to obtain a good match of the experimental µa (λ) of both kidney conditions. The weights used in those reconstructions were estimated using the least squares method, and assuming a total water content of 77% in both kidney tissues, it was possible to calculate the concentrations of the other tissue components. It has been shown that with the development of cancer, the concentrations of proteins, DNA, oxygenated hemoglobin, lipids, and lipofuscin increase, and the concentration of melanin decreases. Future studies based on minimally invasive spectral measurements will allow cancer diagnosis using the proposed approach.

First Insights on the Bioaccessibility and Absorption of Anthocyanins from Edible Flowers: Wild Pansy, Cosmos, and Cornflower.

Edible flowers are regaining interest among both the scientific community and the general population, not only for their appealing sensorial characteristics but also from the growing evidence about their health benefits. Among edible flowers, those that contain anthocyanins are among the most consumed worldwide. However, little is known regarding the bioaccessibility and absorption of their bioactive compounds upon ingestion. The aim of this work was to explore, for the first time, the behavior of anthocyanin-rich extracts from selected edible flowers under different food processing conditions and after ingestion using simulated digestions, as well as their absorption at the intestinal level. Overall, the results showed that the monoglucoside and rutinoside anthocyanin extracts were less stable under different pH, temperature, and time conditions as well as different digestive processes in the gastrointestinal tract. There was a prominent decrease in the free anthocyanin content after the intestinal phase, which was more pronounced for the rutinoside anthocyanin extract (78.41% decrease from the oral phase). In contrast, diglucoside and rutinoside anthocyanin extracts showed the highest absorption efficiencies at the intestinal level, of approximately 5% after 4 h of experiment. Altogether, the current results emphasize the influence of anthocyanins' structural arrangement on both their chemical stability as well as their intestinal absorption. These results bring the first insights about the bioaccessibility and absorption of anthocyanins from wild pansy, cosmos, and cornflower and the potential outcomes of such alternative food sources.

Influence of Chitosan 0.2% in Various Final Cleaning Methods on the Bond Strength of Fiberglass Post to Intrarradicular Dentin.

The purpose of this study was to analyze the influence of Chitosan 0.2% in various final cleaning methods on the bond strength of fiberglass post (FP) to intrarradicular dentin. Ninety bovine incisors were sectioned to obtain root remnants measuring 18 mm in length. The roots were divided: G1: EDTA 17%; G2: EDTA 17% + PUI; G3: EDTA 17% + EA; G4: EDTA 17% + XPF; G5: Chitosan 2%; G6: Chitosan 2% + PUI; G7: Chitosan 2% + EA; G8: Chitosan 2% +XPF. After carrying out the cleaning methods, the posts were installed, and the root was cleaved to generate two disks from each root third. Bond strength values (MPa) obtained from the micro push-out test data were assessed by using Kruskal-Wallis and Dwass-Steel-Critchlow-Fligner tests for multiple comparisons (α = 5%). Differences were observed in the cervical third between G1 and G8 (p = 0.038), G4 and G8 (p = 0.003), G6 and G8 (p = 0.049), and Control and G8 (p = 0.019). The final cleaning method influenced the adhesion strength of cemented FP to intrarradicular dentin. Chitosan 0.2% + XPF positively influenced adhesion strength, with the highest values in the cervical third.

Mortality prediction using medical time series on TBI patients.

BACKGROUND AND OBJECTIVE: Traumatic Brain Injury (TBI) is one of the leading causes of injury-related mortality in the world, with severe cases reaching mortality rates of 30-40%. It is highly heterogeneous both in causes and consequences making more complex the medical interpretation and prognosis. Gathering clinical, demographic, and laboratory data to perform a prognosis requires time and skill in several clinical specialties. Artificial intelligence (AI) methods can take advantage of existing data by performing helpful predictions and guiding physicians toward a better prognosis and, consequently, better healthcare. The objective of this work was to develop learning models and evaluate their capability of predicting the mortality of TBI. The predictive model would allow the early assessment of the more serious cases and scarce medical resources can be pointed toward the patients who need them most. METHODS: Long Short Term Memory (LSTM) and Transformer architectures were tested and compared in performance, coupled with data imbalance, missing data, and feature selection strategies. From the Medical Information Mart for Intensive Care III (MIMIC-III) dataset, a cohort of TBI patients was selected and an analysis of the first 48 hours of multiple time series sequential variables was done to predict hospital mortality. RESULTS: The best performance was obtained with the Transformer architecture, achieving an AUC of 0.907 with the larger group of features and trained with class proportion class weights and binary cross entropy loss. CONCLUSIONS: Using the time series sequential data, LSTM and Transformers proved to be both viable options for predicting TBI hospital mortality in 48 hours after admission. Overall, using sequential deep learning models with time series data to predict TBI mortality is viable and can be used as a helpful indicator of the well-being of patients.

Evaluating the ability of an artificial-intelligence cloud-based platform designed to provide information prior to locoregional therapy for breast cancer in improving patient's satisfaction with therapy: The CINDERELLA trial.

BACKGROUND: Breast cancer therapy improved significantly, allowing for different surgical approaches for the same disease stage, therefore offering patients different aesthetic outcomes with similar locoregional control. The purpose of the CINDERELLA trial is to evaluate an artificial-intelligence (AI) cloud-based platform (CINDERELLA platform) vs the standard approach for patient education prior to therapy. METHODS: A prospective randomized international multicentre trial comparing two methods for patient education prior to therapy. After institutional ethics approval and a written informed consent, patients planned for locoregional treatment will be randomized to the intervention (CINDERELLA platform) or controls. The patients in the intervention arm will use the newly designed web-application (CINDERELLA platform, CINDERELLA APProach) to access the information related to surgery and/or radiotherapy. Using an AI system, the platform will provide the patient with a picture of her own aesthetic outcome resulting from the surgical procedure she chooses, and an objective evaluation of this aesthetic outcome (e.g., good/fair). The control group will have access to the standard approach. The primary objectives of the trial will be i) to examine the differences between the treatment arms with regards to patients' pre-treatment expectations and the final aesthetic outcomes and ii) in the experimental arm only, the agreement of the pre-treatment AI-evaluation (output) and patient's post-therapy self-evaluation. DISCUSSION: The project aims to develop an easy-to-use cost-effective AI-powered tool that improves shared decision-making processes. We assume that the CINDERELLA APProach will lead to higher satisfaction, better psychosocial status, and wellbeing of breast cancer patients, and reduce the need for additional surgeries to improve aesthetic outcome.

Transformers for Energy Forecast.

Forecasting energy consumption models allow for improvements in building performance and reduce energy consumption. Energy efficiency has become a pressing concern in recent years due to the increasing energy demand and concerns over climate change. This paper addresses the energy consumption forecast as a crucial ingredient in the technology to optimize building system operations and identifies energy efficiency upgrades. The work proposes a modified multi-head transformer model focused on multi-variable time series through a learnable weighting feature attention matrix to combine all input variables and forecast building energy consumption properly. The proposed multivariate transformer-based model is compared with two other recurrent neural network models, showing a robust performance while exhibiting a lower mean absolute percentage error. Overall, this paper highlights the superior performance of the modified transformer-based model for the energy consumption forecast in a multivariate step, allowing it to be incorporated in future forecasting tasks, allowing for the tracing of future energy consumption scenarios according to the current building usage, playing a significant role in creating a more sustainable and energy-efficient building usage.

Single Modality vs. Multimodality: What Works Best for Lung Cancer Screening?

In a clinical context, physicians usually take into account information from more than one data modality when making decisions regarding cancer diagnosis and treatment planning. Artificial intelligence-based methods should mimic the clinical method and take into consideration different sources of data that allow a more comprehensive analysis of the patient and, as a consequence, a more accurate diagnosis. Lung cancer evaluation, in particular, can benefit from this approach since this pathology presents high mortality rates due to its late diagnosis. However, many related works make use of a single data source, namely imaging data. Therefore, this work aims to study the prediction of lung cancer when using more than one data modality. The National Lung Screening Trial dataset that contains data from different sources, specifically, computed tomography (CT) scans and clinical data, was used for the study, the development and comparison of single-modality and multimodality models, that may explore the predictive capability of these two types of data to their full potential. A ResNet18 network was trained to classify 3D CT nodule regions of interest (ROI), whereas a random forest algorithm was used to classify the clinical data, with the former achieving an area under the ROC curve (AUC) of 0.7897 and the latter 0.5241. Regarding the multimodality approaches, three strategies, based on intermediate and late fusion, were implemented to combine the information from the 3D CT nodule ROIs and the clinical data. From those, the best model-a fully connected layer that receives as input a combination of clinical data and deep imaging features, given by a ResNet18 inference model-presented an AUC of 0.8021. Lung cancer is a complex disease, characterized by a multitude of biological and physiological phenomena and influenced by multiple factors. It is thus imperative that the models are capable of responding to that need. The results obtained showed that the combination of different types may have the potential to produce more comprehensive analyses of the disease by the models.

Machine learning-based approaches for cancer prediction using microbiome data.

Emerging evidence of the relationship between the microbiome composition and the development of numerous diseases, including cancer, has led to an increasing interest in the study of the human microbiome. Technological breakthroughs regarding DNA sequencing methods propelled microbiome studies with a large number of samples, which called for the necessity of more sophisticated data-analytical tools to analyze this complex relationship. The aim of this work was to develop a machine learning-based approach to distinguish the type of cancer based on the analysis of the tissue-specific microbial information, assessing the human microbiome as valuable predictive information for cancer identification. For this purpose, Random Forest algorithms were trained for the classification of five types of cancer-head and neck, esophageal, stomach, colon, and rectum cancers-with samples provided by The Cancer Microbiome Atlas database. One versus all and multi-class classification studies were conducted to evaluate the discriminative capability of the microbial data across increasing levels of cancer site specificity, with results showing a progressive rise in difficulty for accurate sample classification. Random Forest models achieved promising performances when predicting head and neck, stomach, and colon cancer cases, with the latter returning accuracy scores above 90% across the different studies conducted. However, there was also an increased difficulty when discriminating esophageal and rectum cancers, failing to differentiate with adequate results rectum from colon cancer cases, and esophageal from head and neck and stomach cancers. These results point to the fact that anatomically adjacent cancers can be more complex to identify due to microbial similarities. Despite the limitations, microbiome data analysis using machine learning may advance novel strategies to improve cancer detection and prevention, and decrease disease burden.

Influence of different agitation techniques on bacterial reduction in curved root canals.

The aim of this study was to evaluate the influence of agitation techniques on bacterial reduction in curved root canals. Eighty human mandibular molars were prepared, inoculated with Enterococcus faecalis and incubated at 37°C for 60 days. Then, specimens were randomly separated into two test groups (n = 36) and two control groups (n = 04) according to agitation technique: Passive ultrasonic irrigation (PUI, Irrisonic) and XP-endo Finisher (XPF). Microbial samples were collected before and after instrumentation and after final agitation using sterile paper points. Bacterial growth was analysed by turbidity of culture medium and UV spectrophotometry. The Wilcoxon rank test was used for the paired analysis, while the Mann-Whitney U-test was used for the non-paired analysis. The samples collected after final agitation were significantly different between test groups (p < 0.05). Bacterial reduction was greater in the PUI than in the XPF (p < 0.05) group. The irrigant agitation provided significant bacterial reduction. The use of the PUI showed better results.

Acurácia da detecção do antígeno galactomanana no soro para o diagnóstico precoce de aspergilose pulmonar invasiva: uma revisão rápida / Accuracy of serum galactomannan antigen detection for the early diagnosis of invasive pulmonary aspergillosis: a rapid review

Tecnologia: Detecção do antígeno galactomanana no soro. Contexto: A aspergilose pulmonar invasiva (API) é uma infecção fúngica oportunista de grande risco para pacientes imunocomprometidos. A detecção do antígeno galactomanana no soro por meio de um imunoensaio (ELISA) pode ser um teste não invasivo que auxilie no diagnóstico precoce da doença nestes pacientes. Objetivo: Avaliar a acurácia da detecção do antígeno galactomana no soro para o diagnóstico precoce de aspergilose pulmonar invasiva. Métodos: Revisão rápida sistematizada sobre acurácia de diagnóstico. As bases de dados utilizadas na pesquisa foram: PUBMED, EMBASE, SCOPUS, BVS e Cochrane Library. A avaliação da qualidade metodológica dos estudos incluídos foi realizada por meio da ferramenta AMSTAR-2. Resultados: Foram selecionadas três revisões sistemáticas que atendiam aos critérios de elegibilidade com as quais foi realizada uma análise descritiva dos dados encontrados. A avaliação da qualidade metodológica demonstrou que duas das revisões sistemáticas (RS) apresentaram qualidade criticamente baixa e uma das RS apresentou qualidade alta. Conclusão: A detecção da galactomanana sérica por ELISA pode ser um teste auxiliar no diagnóstico de API, entretanto, possui várias limitações e deve ser utilizado juntamente com outros critérios diagnósticos do consenso do EORTC/MSG. Novas pesquisas devem ser fomentadas para avaliar a utilização do teste no tempo do diagnóstico e no monitoramento da API

Technology: Detection of galactomannan antigen in serum. Background: Invasive pulmonary aspergillosis (IPA) is an opportunistic fungal infection of serious risk for immunocompromised patients. Detection of galactomannan antigen in serum by immunoassay (ELISA) could be a noninvasive test that contributes to the early diagnosis of the disease in this group of patients. Objective: To evaluate the accuracy of serum galactomannan antigen detection for the early diagnosis of invasive pulmonary aspergillosis. Methods: Rapid review of diagnostic accuracy. Databases used in the search were: PUBMED, EMBASE, SCOPUS, BVS, and Cochrane Library. The methodological quality of the included studies was assessed using the AMSTAR-2 tool. Results: Three systematic reviews that satisfied the eligibility criteria were selected, and a descriptive analysis of the data found was performed. The methodological quality assessment showed that two of the systematic reviews (SR) presented critically low quality, and one of the SR presented high quality. Conclusion: Detection of serum galactomannan by ELISA may be a valuable test for diagnosing IPA; however, it has a series of limitations and should be used in conjunction with other diagnostic criteria of the EORTC/MSG consensus. Further research should be encouraged to evaluate the use of this assay, considering the time to diagnosis and IPA monitoring

Behind the Scenes of Anthocyanins-From the Health Benefits to Potential Applications in Food, Pharmaceutical and Cosmetic Fields.

Anthocyanins are widespread and biologically active water-soluble phenolic pigments responsible for a wide range of vivid colours, from red (acidic conditions) to purplish blue (basic conditions), present in fruits, vegetables, and coloured grains. The pigments' stability and colours are influenced mainly by pH but also by structure, temperature, and light. The colour-stabilizing mechanisms of plants are determined by inter- and intramolecular co-pigmentation and metal complexation, driven by van der Waals, π-π stacking, hydrogen bonding, and metal-ligand interactions. This group of flavonoids is well-known to have potent anti-inflammatory and antioxidant effects, which explains the biological effects associated with them. Therefore, this review provides an overview of the role of anthocyanins as natural colorants, showing they are less harmful than conventional colorants, with several technological potential applications in different industrial fields, namely in the textile and food industries, as well as in the development of photosensitizers for dye-sensitized solar cells, as new photosensitizers in photodynamic therapy, pharmaceuticals, and in the cosmetic industry, mainly on the formulation of skin care formulations, sunscreen filters, nail colorants, skin & hair cleansing products, amongst others. In addition, we will unveil some of the latest studies about the health benefits of anthocyanins, mainly focusing on the protection against the most prevalent human diseases mediated by oxidative stress, namely cardiovascular and neurodegenerative diseases, cancer, and diabetes. The contribution of anthocyanins to visual health is also very relevant and will be briefly explored.

Preparation of nanoliposomes loaded with anthocyanins from grape skin extracts: stability, gastric absorption and antiproliferative properties.

Anthocyanin nanoliposomes (ANLs) were produced by a thin film ultrasonic dispersion method to improve the stability and bioavailability of anthocyanins (ACNs) obtained from grape skin extracts. The preparation parameters were predicted to be a soy lecithin to cholesterol ratio of 80 : 19 (w/w, 2 mg of ACNs) under ultrasonication at 120 W for 3.12 min by the response surface methodology. Under the optimal conditions, the fabricated ANLs presented an encapsulation efficiency of 40.1% with an average particle size of 117 nm, a PDI of 0.254 and a ζ-potential of 8.56 mV. The stability of ACNs was improved by nanoliposome encapsulation under various temperature and light conditions. Moreover, a MKN-28 (stomach) barrier model was established to evaluate the cellular transport of ACNs before and after nanoliposome encapsulation. HPLC-DAD/MS analyses demonstrated that ACNs obtained from grape skin extracts mainly consisted of five monomers. After 180 min of transportation, peonidin-3-5-diglucoside and malvidin-3-5-diglucoside (two representative monomers) present in ANLs (0.5 mg mL-1) showed the maximum transport efficiencies of 17.25 ± 1.62% and 18.94 ± 1.05%, respectively. However, their maximum transport efficiencies were 11.68 ± 1.01% and 15.33 ± 1.24%, respectively, existing in ACNs (non-encapsulated form, 0.5 mg mL-1). Furthermore, the antiproliferative properties of ANLs were assessed in two cancer cell lines MKN-28 and HepG-2 (liver). The ANLs presented more effective antiproliferative effects towards MKN-28 than the HepG-2 cell line. This study provides theories and a practice foundation for further application of ACNs as nutraceutical and functional foods.

The botanical profile, phytochemistry, biological activities and protected-delivery systems for purple sweet potato (Ipomoea batatas (L.) Lam.): An up-to-date review.

Purple sweet potato (PSP) is an important economic crop in many countries, as a staple food and a source of bioactive compounds, which has attracted considerable attention. This review provides an up-to-date summary and discusses the available literature concerning PSP. Different issues, including its bioactive compounds, health effects and various efficient encapsulation strategies for PSP powders, extracts or individual substance are covered in detail, along with its utilization. In addition to the valuable nutritional composition, more than 135 bioactive compounds have been isolated and identified from these plants so far. Among the plenty of constituents, polysaccharides and flavonoids are the focus of attention and exhibit various biological activities.Additionally, protected-delivery systems are strongly proposed to shelter the bioactive compounds providing a better stability and improved pharmacological activities. Normally, PSP roots are the most attractive part to human because of their economic value. Even though PSP anthocyanins are the focus of researchers and industrial due to their attractive color and wide range of biological activities, PSP starch and protein also have wide applications in foods and nonfoods industries. However, the exploitation of PSP considering comprehensive utilization of various compounds, such as starch, non-starch polysaccharides, protein, and bioactive compounds should be considered.

Impact of a Purified Blueberry Extract on In Vitro Probiotic Mucin-Adhesion and Its Effect on Probiotic/Intestinal Pathogen Systems.

Several arguments have been made to substantiate the need for natural antimicrobials for the food industry. With blueberry extracts, the most compelling are both their healthy connotation and the possibility of obtaining a multipurpose solution that can be an antioxidant, colorant, and antimicrobial. From an antimicrobial perspective, as blueberry/anthocyanin-rich extracts have been associated with a capacity to inhibit harmful bacteria while causing little to no inhibition on potential probiotic microorganisms, the study of potential benefits that come from synergies between the extract and probiotics may be of particular interest. Therefore, the present work aimed to evaluate the effect of an anthocyanin-rich extract on the adhesion of five different probiotics as well as their effect on the probiotics' capacity to compete with or block pathogen adhesion to a mucin/BSA-treated surface. The results showed that, despite some loss of probiotic adhesion, the combined presence of extract and probiotic is more effective in reducing the overall amount of adhered viable pathogen cells than the PROBIOTIC alone, regardless of the probiotic/pathogen system considered. Furthermore, in some instances, the combination of the extract with Bifidobacterium animalis Bo allowed for almost complete inhibition of pathogen adhesion.

Synthesizing 3D Lung CT scans with Generative Adversarial Networks.

In the healthcare domain, datasets are often private and lack large amounts of samples, making it difficult to cope with the inherent patient data heterogeneity. As an attempt to mitigate data scarcity, generative models are being used due to their ability to produce new data, using a dataset as a reference. However, synthesis studies often rely on a 2D representation of data, a seriously limited form of information when it comes to lung computed tomography scans where, for example, pathologies like nodules can manifest anywhere in the organ. Here, we develop a 3D Progressive Growing Generative Adversarial Network capable of generating thoracic CT volumes at a resolution of 1283, and analyze the model outputs through a quantitative metric (3D Muli-Scale Structural Similarity) and a Visual Turing Test. Clinical relevance - This paper is a novel application of the 3D PGGAN model to synthesize CT lung scans. This preliminary study focuses on synthesizing the entire volume of the lung rather than just the lung nodules. The synthesized data represent an attempt to mitigate data scarcity which is one of the major limitations to create learning models with good generalization in healthcare.

On the way for the best imaging features from CT images to predict EGFR Mutation Status in Lung Cancer.

Artificial Intelligence-based tools have shown promising results to help clinicians in diagnosis tasks. Radio-genomics would aid in the genotype characterization using information from radiologic images. The prediction of the mutations status of main oncogenes associated with lung cancer will help the clinicians to have a more accurate diagnosis and a personalized treatment plan, decreasing the need to use the biopsy. In this work, novel and objective features were extracted from the lung that contained the nodule, and several machine learning methods were combined with feature selection techniques to select the best approach to predict the EGFR mutation status in lung cancer CT images. An AUC of 0.756 ± 0.055 was obtained using a logistic regression and independent component analysis as feature selector, supporting the hypothesis that CT images can capture pathophysiological information with great value for clinical assessment and personalized medicine of lung cancer. Clinical Relevance - Radiogenomic approaches could be an interesting help for lung cancer characterization. This work represents a preliminary study for the development of computer-aided decision systems to provide a more accurate and fast characterization of lung cancer which is fundamental for an adequate treatment plan for lung cancer patients.

Unsupervised Approach for Malignancy Assessment of Lung Nodules in Computed Tomography Scans Using Radiomic Features.

Lung cancer is the leading cause of cancer death worldwide. Early low-dose computed tomography (CT) screening can decrease its mortality rate and computer-aided diagnoses systems may make these screenings more accessible. Radiomic features and supervised machine learning have traditionally been employed in these systems. Contrary to supervised methods, unsupervised learning techniques do not require large amounts of annotated data which are labor-intensive to gather and long training times. Therefore, recent approaches have used unsupervised methods, such as clustering, to improve the performance of supervised models. However, an analysis of purely unsupervised methods for malignancy prediction of lung nodules from CT images has not been performed. This work studies nodule malignancy in the LIDC-IDRI image collection of chest CT scans using established radiomic features and unsupervised learning methods based on k-Means, Spectral Clustering, and Gaussian Mixture clustering. All tested methods resulted in clusters of high homogeneity malignancy. Results suggest convex feature distributions and well-separated feature subspaces associated with different diagnoses. Furthermore, diagnosis uncertainty may be explained by common characteristics captured by radiomic features. The k-Means and Gaussian Mixture models are able to generalize to unseen data, achieving a balanced accuracy of 87.23% and 86.96% when inference was tested. These results motivate the usage of unsupervised approaches for malignancy prediction of lung nodules, such as cluster-then-label models. Clinical Relevance- Unsupervised clustering of radiomic features of lung nodules in chest CT scans can differentiate between malignant and benign cases and reflects experts' diagnosis uncertainty.

A Random Forest-based Classifier for MYCN Status Prediction in Neuroblastoma using CT Images.

Neuroblastoma (NB) is the most common extracranial solid tumor in childhood. Genomic amplification of MYCN is associated with poor outcomes and is detected in 16% of all NB cases. CT scans and MRI are the imaging techniques recommended for diagnosis and disease staging. The assessment of imaging features such as tumor volume, shape, and local extension represent relevant prognostic information. Radiogenomics have shown powerful results in the assessment of the genotype based on imaging findings automatically extracted from medical images. In this work, random forest was used to classify the MYCN amplification using radiomic features extracted from CT slices in a population of 46 NB patients. The learning model showed an area under the curve (AUC) of 0.85 ± 0.13, suggesting that radiomic-based methodologies might be helpful in the extraction of information that is not accessible by human naked eyes but could aid the clinicians on the diagnosis and treatment plan definition. Clinical relevance - This approach represents a random forest-based model to predict the MYCN amplification in NB patients that could give a faster, earlier, and repeatable analysis of the tumor along the time.

Learning Models for Traumatic Brain Injury Mortality Prediction on Pediatric Electronic Health Records.

Background: Traumatic Brain Injury (TBI) is one of the leading causes of injury related mortality in the world, with severe cases reaching mortality rates of 30-40%. It is highly heterogeneous both in causes and consequences, complicating medical interpretation and prognosis. Gathering clinical, demographic, and laboratory data to perform a prognosis requires time and skill in several clinical specialties. Machine learning (ML) methods can take advantage of the data and guide physicians toward a better prognosis and, consequently, better healthcare. The objective of this study was to develop and test a wide range of machine learning models and evaluate their capability of predicting mortality of TBI, at hospital discharge, while assessing the similarity between the predictive value of the data and clinical significance. Methods: The used dataset is the Hackathon Pediatric Traumatic Brain Injury (HPTBI) dataset, composed of electronic health records containing clinical annotations and demographic data of 300 patients. Four different classification models were tested, either with or without feature selection. For each combination of the classification model and feature selection method, the area under the receiver operator curve (ROC-AUC), balanced accuracy, precision, and recall were calculated. Results: Methods based on decision trees perform better when using all features (Random Forest, AUC = 0.86 and XGBoost, AUC = 0.91) but other models require prior feature selection to obtain the best results (k-Nearest Neighbors, AUC = 0.90 and Artificial Neural Networks, AUC = 0.84). Additionally, Random Forest and XGBoost allow assessing the feature's importance, which could give insights for future strategies on the clinical routine. Conclusion: Predictive capability depends greatly on the combination of model and feature selection methods used but, overall, ML models showed a very good performance in mortality prediction for TBI. The feature importance results indicate that predictive value is not directly related to clinical significance.