Exploring the use of association rules in random forest for predicting heart disease

Heart disease is one of the most dangerous diseases in the world. People with these diseases, most of them end up losing their lives. Therefore, machine learning algorithms have proven to be useful in this sense to help decision-making and prediction from the large amount of data generated by the healthcare sector. In this work, we have proposed a novel method that allows increasing the performance of the classical random forest technique so that this technique can be used for the prediction of heart disease with its better performance. We used in this study other classifiers such as classical random forest, support vector machine, decision tree, Naïve Bayes, and XGBoost. This work was done in the heart dataset Cleveland. According to the experimental results, the accuracy of the proposed model is better than that of other classifiers with 83.5%.This study contributed to the optimization of the random forest technique as well as gave solid knowledge of the formation of this technique.

Fructan supplementation of senior cats affects stool metabolite concentrations and fecal microbiota concentrations, but not nitrogen partitioning in excreta

Fructan supplementation of a commercially available canned cat food was evaluated using senior (≥ 9 yr) cats to assess nitrogen (N) partitioning in excreta and stool metabolite and microbiota concentrations. Oligofructose (OF) or SynergyC (OF+IN) were added to the diet individually at 1% (dry weight basis). Cats were acclimated to the control diet for 7 d and then were randomly assigned to 1 of 3 treatment groups for 21 d (n = 6). Feces and urine were collected on d 22 through 28. No differences were observed in food intake; fecal output, DM percentage, score, pH, or short- or branched-chain fatty acids, fecal and urinary ammonia output, urinary felinine concentrations, or N retention. Supplemental OF+IN tended to decrease N digestibility (P = 0.102) and Bifidobacteria spp. (P = 0.073) and decrease fecal indole (P < 0.05), tyramine (P < 0.05), and Escherichia coli (P < 0.05) concentrations. Both fructan-supplemented treatments decreased (P < 0.05) fecal histamine concentrations. The tendency to a lower apparent N digestibility was likely due to increased colonic microbial protein synthesis of fructan-supplemented cats. Fructan supplementation may benefit senior cats as it modulates stool odor-forming compounds and decreases some protein catabolites and pathogenic gut microbiota concentrations without affecting N retention.

Adaptation of healthy adult cats to select dietary fibers in vivo affects gas and short-chain fatty acid production from fiber fermentation in vitro

Nine young adult (1.73 ± 0.03 yr) male cats were used to determine the effects of microbial adaptation to select dietary fiber sources on changes in pH in vitro and on total and hydrogen gas, short-chain fatty acid (SCFA), and branched-chain fatty acid (BCFA) production. Cats were adapted to diets containing 4% cellulose, fructooligosaccharides (FOS), or pectin for 30 d before fecal sampling. Each cat was used as a single donor, and fecal inoculum was reacted with each of the aforementioned fiber substrates. Adaptation to dietary FOS resulted in a greater change in pH when exposed to FOS than pectin (adaptation × substrate, P < 0.001). When exposed to the FOS substrate, adaptation to dietary FOS or pectin increased hydrogen gas production (adaptation × substrate, P = 0.021). Adaptation to dietary FOS increased acetate and total SCFA production when exposed to FOS substrate in vitro (adaptation × substrate, P = 0.001). When exposed to the FOS substrate, propionate production tended to increase with adaptation to dietary cellulose (adaptation × substrate, P = 0.060). The BCFA + valerate tended to decrease with adaptation to dietary FOS when exposed to FOS substrate in vitro (adaptation × substrate, P = 0.092). Fructooligosaccharides resulted in the greatest change in pH and production of total gas (P < 0.001), hydrogen gas (P < 0.001), acetate (P < 0.001), propionate (P < 0.001), butyrate (P < 0.001), total SCFA (P < 0.001), and total BCFA + valerate production (P < 0.001). Adaptation to the FOS or pectin diet increased production of hydrogen gas with FOS and pectin substrates. Adaptation to pectin increased (P = 0.033) total gas production with FOS and pectin substrates. Overall, adaptation to either FOS or pectin led to greater SCFA and gas production, but adaptation to FOS resulted in the greatest effect overall.

Dietary cellulose, fructooligosaccharides, and pectin modify fecal protein catabolites and microbial populations in adult cats

Twelve young adult (1.7 +/- 0.1 yr) male cats were used in a replicated 3 x 3 Latin square design to determine the effects of fiber type on nutrient digestibility, fermentative end products, and fecal microbial populations. Three diets containing 4% cellulose, fructooligosaccharides (FOS), or pectin were evaluated. Feces were scored based on the 5-point system: 1 being hard, dry pellets, and 5 being watery liquid that can be poured. No differences were observed (P > 0.100) in intake of DM, OM, CP, or acid-hydrolyzed fat; DM or OM digestibility; or fecal pH, DM%, output on an as-is or DM basis, or concentrations of histamine or phenylalanine. Crude protein and fat digestibility decreased (P = 0.079 and 0.001, respectively) in response to supplementation with pectin compared with cellulose. Both FOS and pectin supplementation resulted in increased fecal scores (P < 0.001) and concentrations of ammonia (P = 0.003) and 4-methyl phenol (P = 0.003). Fecal indole concentrations increased (P = 0.049) when cats were supplemented with FOS. Fecal acetate (P = 0.030), propionate (P = 0.035), and total short-chain fatty acid (P = 0.016) concentrations increased in pectin-supplemented cats. Fecal butyrate (P = 0.010), isobutyrate (P = 0.011), isovalerate (P = 0.012), valerate (P = 0.026), and total branched-chain fatty acids + valerate (P = 0.008) concentrations increased with supplementation of FOS and pectin. Fecal cadaverine (P < 0.001) and tryptamine (P < 0.001) concentrations increased with supplementation of FOS and pectin. Fecal tyramine concentrations decreased (P = 0.039) in FOS-supplemented cats, whereas spermidine concentrations increased (P < 0.001) in pectin-supplemented cats. Whereas fecal concentrations of putrescine (P < 0.001) and total biogenic amines (P < 0.001) increased with FOS and pectin, the concentrations of these compounds were increased (P < 0.001) in cats supplemented with pectin. Fecal Bifidobacterium spp. concentrations increased (P = 0.006) and Escherichia coli concentrations decreased (P < 0.001) in FOS-supplemented cats. Fecal concentrations of Clostridium perfringens (P < 0.001), E. coli (P < 0.001), and Lactobacillus spp. (P = 0.030) also increased in pectin-supplemented cats. In addition to increasing populations of protein-fermenting microbiota, pectin increased production of fermentative end products associated with carbohydrate compared with protein fermentation. Pectin and FOS may be useful fiber sources in promoting intestinal health of the cat.

Eosinophilic cystitis: CT findings

Eosinophilic cystitis (EC) is a rare form of bladder inflammation of uncertain etiology, characterized by eosinophilic inflammation of the bladder. Presenting clinical symptoms include hematuria, dysuria, frequency, and suprapubic pain. To our knowledge, there have been only two other reports of computed tomographic (CT) scan findings of EC. In both cases, the findings were thought to represent invasive bladder cancer. We present a case report of biopsy-proven EC with previously unreported CT scan findings.