Advanced Computing for Cardiovascular Disease Prediction

Developing a predictive model for detecting cardiovascular diseases (CVDs) is crucial due to its high global fatality rate. With the advancements in artificial intelligence, the availability of large-scale data, and increased access to computational capability, it is feasible to create robust models that can detect CVDs with high precision. This study aims to provide a promising method for early diagnosis by employing various machine learning and deep learning techniques, including logistic regression, decision trees, random forest classifier, extreme gradient boosting (XGBoost), and a sequential model from Keras. Our evaluation identifies the random forest classifier as the most effective model, achieving an accuracy of 0.91, surpassing other machine learning and deep learning approaches. Close behind are XGBoost (accuracy: 0.90), decision tree (accuracy: 0.86), and logistic regression (accuracy: 0.70). Additionally, our deep learning sequential model demonstrates promising classification performance, with an accuracy of 0.80 and a loss of 0.425 on the validation set. These findings underscore the potential of machine learning and deep learning methodologies in advancing cardiovascular disease prediction and management strategies.

Contrasting lipid biomarkers in mountain rivers in the Nepal Himalayas: Organic matter characteristics and contribution to the fluvial carbon pool

The Nepal Himalayas is the source of many glacial and spring-fed river systems crisscrossing the mountainous terrain.There is an increasing recognition of small mountain rivers(SMRs)to have a significant combined export of dissolved and particulate organic carbon to the global carbon flux.We analyzed fluvial sediments from two SMRs and compared the results with two large mountain rivers(LMRs)in Nepal.We investigated the organic matter(OM),its compositional variability,and seasonal export using a suite of lipid biomarkers,namely n-alkanes,n-alkanoic acids,n-alkanols,and sterols.The SMRs indicated a similarity in lipid distribution and were affected by a strong seasonal variability.The LMRs showed a distinct contrast in the distribution of lipids in suspended sediments.Bedload sediments in SMRs were derived from diverse sources with weak terrigenous dominance all-year-round compared to the suspended load.Functional lipids(n-alkanoic acids and n-alkanols)were the major constituents in SMR sediments,indicating better preservation.In contrast,n-alkane concentration dominated over other fractions in suspended sediments retrieved from LMRs.The biomarker trends differentiate SMRs from LMRs with lower transformed/degraded OM in SMRs.A common observation was the strong presence of even carbon compounds in short-chain n-alkanes in SMR bedload sediments and their predominance in suspended sediments in LMRs.Such an unusual trend is attributed to specific biomarker sources from the catchment and ongoing processes in fluvial systems.Topsoil colonized by fungal species under moist acidic conditions and autochthonous bacteria contributes to the organic matter pool in shallow SMRs.In LMRs,the contribution from thermally mature sedimentary hydrocarbons and the diagenetic reduction of nalkanoic acids to n-alkanes are additional contributors to the allochthonous carbon pool.The differences in lipid concentrations,their distribution,seasonality,and the size of rivers suggest differential preservation/degradation of the organic matter pool and their importance in contributing to the carbon budget.

Hydro-Chemistry of Gokyo Valley, Sagarmatha (Everest) National Park, Nepal

This research work records the baseline inventory of selected ions in running waters of Gokyo Valley, Nepal and tries to establish relationship of these ions’ concentration with elevation gradient. Among these selected hydro-chemical parameters, the average concentration of Ca2+, Mg2+, and total silica were determined to be 3.25 (±2.02), 0.64 (±0.47), 207.2 (±423.01) and 0.36 (±0.17) mg/L respectively. Concentration of these parameters was low compared to earlier studies except that could be justified too since the sampling was performed only once in drier period with low atmospheric input of acid protons required for weathering process. Moreover, Pearson Product-moment Correlation Coefficients (r) for elevation gradient and concentration of Ca2+, Mg2+, and total silica were determined to be +0.350, +0.145, +0.101 and &#450.275 respectively. Such weak relationship between these parameters and elevation gradient suggests that the concentration of these ions is related to weathering phenomena, glacier and watershed characteristics and snow cover dynamics in Gokyo Valley.