Liver cancer is the sixth most frequently diagnosed cancer and, particularly, Hepatocellular Carcinoma (HCC) represents more than 90% of primary liver cancers. Clinicians assess each patient's treatment on the basis of evidence-based medicine, which may not always apply to a specific patient, given the biological variability among individuals. Over the years, and for the particular case of Hepatocellular Carcinoma, some research studies have been developing strategies for assisting clinicians in decision making, using computational methods (e.g. machine learning techniques) to extract knowledge from the clinical data. However, these studies have some limitations that have not yet been addressed: some do not focus entirely on Hepatocellular Carcinoma patients, others have strict application boundaries, and none considers the heterogeneity between patients nor the presence of missing data, a common drawback in healthcare contexts. In this work, a real complex Hepatocellular Carcinoma database composed of heterogeneous clinical features is studied. We propose a new cluster-based oversampling approach robust to small and imbalanced datasets, which accounts for the heterogeneity of patients with Hepatocellular Carcinoma. The preprocessing procedures of this work are based on data imputation considering appropriate distance metrics for both heterogeneous and missing data (HEOM) and clustering studies to assess the underlying patient groups in the studied dataset (K-means). The final approach is applied in order to diminish the impact of underlying patient profiles with reduced sizes on survival prediction. It is based on K-means clustering and the SMOTE algorithm to build a representative dataset and use it as training example for different machine learning procedures (logistic regression and neural networks). The results are evaluated in terms of survival prediction and compared across baseline approaches that do not consider clustering and/or oversampling using the Friedman rank test. Our proposed methodology coupled with neural networks outperformed all others, suggesting an improvement over the classical approaches currently used in Hepatocellular Carcinoma prediction models.

BackgroundFast and accurate chest pain risk stratification in the emergency department (ED) is critical. The HEART score predicts the short-term incidence of major adverse cardiac events (MACE) in this population, dividing it in three risk categories. We aimed to describe the population with chest pain, to characterize the subgroup of patients with acute coronary syndrome (ACS) and to assess the prognostic value of Manchester triage system and of HEART score.MethodsRetrospective observational study including patients admitted to the ED of a tertiary hospital with chest pain as the presenting symptom. The primary outcome was a composite of all-cause mortality, myocardial infarction or unscheduled revascularization at 6 weeks.ResultsWe enrolled 233 patients (age 58 ± 19; 55.4 % males). The most common final diagnosis was non-specific chest pain (n = 86, 36.9 %), followed by ACS (n = 22, 9.4 %). Male gender, smoking and chronic kidney disease were associated with higher risk of ACS. According to Manchester triage system, chest pain patients stratified with red or orange priority had a higher incidence of ACS (16.5 % vs. 3.8 %, p = 0.006). The application of HEART score showed that most patients were in low risk category (56.3 %). The six-week incidence of MACE in each category was 2 %, 15.6 % and 76.9 % (p < 0.001). HEART score accurately predicted the short-term incidence of MACE in chest pain patients (c-statistic 0.880; 95 % CI, 0.807–0.950, p < 0.001).ConclusionsChest pain patients have very different levels of severity and the discriminatory power of Manchester triage system should be used in the assessment of this population. The HEART score seems to be an effective tool for risk stratification in the ED.

IntroductionTendon injury is a major cause of lameness and decreased performance in athletic equines. Various therapies for tendonitis have been described; however, none of these therapies results in complete tissue regeneration, and the injury recurrence rate is high even after long recovery periods involving rest and physiotherapy.MethodsA lesion was induced with collagenase gel in the superficial digital flexor tendon in the center portion of the metacarpal region of eight equines of mixed breed. After two weeks, the lesions of the animals in the treated and control groups were treated through the intralesional administration of mesenchymal stem cells derived from adipose tissue (adMSCs) suspended in platelet concentrate (PC) and with phosphate buffered saline (PBS), respectively. Serial ultrasound analyses were performed every two weeks. After 16 weeks of therapy, a biopsy was performed for histopathological, immunohistochemical and gene expression (type I collagen (COL1A1), type III collagen (COL3A1), tenascin-C (TNC), tenomodulin (TNMD), and scleraxis (SCX)) analyses.ResultsDifferences in the ultrasound and histopathological analyses were observed between the groups. Improved results were reported in the group treated with adMSCs suspended in PC. There was no difference in the gene expression levels observed after the different treatments. The main results observed from the histopathological evaluation of the treated group were as follows: a prevention of the progression of the lesion, a greater organization of collagen fibers, and a decreased inflammatory infiltrate. A lack of progression of the lesion area and its percentage was observed in the ultrasound image, and increased blood flow was measured by Power Doppler.ConclusionsThe use of adMSCs combined with PC for the therapy of experimentally induced tendonitis prevented the progression of the tendon lesion, as observed in the ultrasound examination, and resulted in a greater organization and decreased inflammation, as observed in the histopathological evaluation. These data demonstrate the therapeutic potential of this therapy for the treatment of equine tendonitis.

No abstract

Superficial digital flexor tendon lesion is an important cause of lameness in equine athletes. Although numerous treatments have been described, few are effective at promoting significant improvement in the quality of the extracellular matrix. Therefore, great potential remains for recurrence and in certain cases, an abrupt end to the horse's athletic career. Recently, several experiments have focused on the therapeutic potential of mesenchymal stem cells (MSCs) in cases of tendon lesions. This study aimed to evaluate the effect of adipose tissue-derived MSCs in the treatment of induced tendinitis of the superficial digital flexor tendon in horses by clinical, ultrasonographic, histopathological, and immunochemical analyses. Tendinitis was induced in both thoracic limbs of eight mares by administration of collagenase solution and adipose tissue was collected from the tail base for MSCs isolation and expansion, which were used during cellular therapy on only one limb 30 days after lesion induction. No differences occurred between the groups regarding the clinical and ultrasonographic analyses; however, histopathological evaluation revealed a significant improvement in tendon fiber organization and diminished inflammatory infiltrate, whereas immunohistochemical analysis showed increased expression of type I collagen in the treated group as compared with controls. The cellular therapy model implanted in this experiment promoted increased perivascular inflammatory infiltrate, fibroblastic density, neovascularization, and qualitative healing improvement of tendon extracellular matrix, in terms of fiber orientation and type I/III collagen ratio; moreover, it was considered to be a safe and viable process.