Multi-class classification algorithms for the diagnosis of anemia in an outpatient clinical setting.

Anemia is one of the most pressing public health issues in the world with iron deficiency a major public health issue worldwide. The highest prevalence of anemia is in developing countries. The complete blood count is a blood test used to diagnose the prevalence of anemia. While earlier studies have framed the problem of diagnosis as a binary classification problem, this paper frames it as a multi class (three classes) classification problem with mild, moderate and severe classes. The three classes for the anemia classification (mild, moderate, severe) are so chosen as the world health organization (WHO) guidelines formalize this categorization based on the Haemoglobin (HGB) values of the chosen sample of patients in the Complete Blood Count (CBC) patient data set. Complete blood count test data was collected in an outpatient clinical setting in India. We used Feature selection with Majority voting to identify the key attributes in the input patient data set. In addition, since the original data set was imbalanced we used Synthetic Minority Oversampling Technique (SMOTE) to balance the data set. Four data sets including the original data set were used to perform the data experiments. Six standard machine learning algorithms were utilised to test our four data sets, performing multi class classification. Benchmarking these algorithms was performed and tabulated using both10 fold cross validation and hold out methods. The experimental results indicated that multilayer perceptron network was predominantly giving good recall values across mild and moderate class which are early and middle stages of the disease. With a good prediction model at early stages, medical intervention can provide preventive measure from further deterioration into severe stage or recommend the use of supplements to overcome this problem.

Adenine based molecular junction as biosensor for detection of toxic phosgene gas.

The possibility of adsorption of toxic phosgene gas (COCl2) molecule on one of the nucleobase of DNA-adenine-has been analyzed using the first principle calculations based on density function theory. In accordance with the geometry of the nucleobase, two possible positions have been considered for effective adsorption of gas molecule. The calculations performed on adsorption energies suggest that the gas molecule is able to physisorb at both the considered positions with negligibly small values of charge transfer. The in-depth analysis of electron charge densities depicts that there is no orbital overlapping between the gas molecule and adenine. We observe a significant variation of transport properties of adenine-based molecular junction on adsorption of phosgene molecule while calculation the transport parameters at both the equilibrium as well as non-equilibrium. Also, the variation of HOMO-LUMO gap of adenine molecule on adsorption of phosgene leads to alteration of current and voltage, thus implying that adenine-based sensor can be effectively utilized to sense the presence of phosgene gas in a given environment. Small adsorption energies and recovery time suggest that the rate of desorption of phosgene is very high; thus, the proposed adenine sensor can be effectively used as a highly stable and selective reusable sensor.

A pursuit to design highly sensitive fullerene-based sensors: adsorption and dissociation phenomenon of toxic sulfur gases on B40 fullerene.

The adsorption phenomenon of toxic sulfur gases namely H2S and SO2 on B40 fullerene is scrutinized utilizing density functional theory-non-equilibrium Green's function (DFT-NEGF) regime. Adsorption of gas molecules is considered at both the hexagonal and heptagonal rings of the fullerene and adsorption energies, charge transfer, electron charge densities, density of states, transmission spectra, molecular energy spectra; Eigen states, HOMO-LUMO gap, current voltage curve, and differential conductance are premeditated. It is inferred that H2S molecule is physisorbed on the heptagonal ring of the fullerene while it is dissociative-chemisorbed on the hexagonal ring. SO2 dissociates into SO and O species on adsorption on both the hexagonal and heptagonal rings. From the transmission spectra and DOS analysis, LUMO dominant transmission is noticed in all the devices except the one formed with heptagonal ring adsorption of H2S which favors HOMO-dominated transmission. From the I-V curve and differential conductance investigation, different conductance values are noticed for all the junctions, thus proving that B40 is an efficient material to be engaged in sensing toxic sulfur gases.

Comparative evaluation of NANO transport properties for DNA nucleobase based molecular junction devices.

The feasibility of electron transport conduction through a guanine base of DNA was investigated and then compared with another component of DNA, i.e., cytosine. A mathematical approach based on the jellium model using non-equilibrium Green's function combined with semi empirical extended Huckel theory was applied using the Atomistik Tool Kit. This was further used to measure significant transport parameters such as current, conductance, transmission spectra and the HOMO-LUMO gap of the suggested molecular system. An important revelation from our research work is that the cytosine-based molecular device exhibits metallic behavior with current ranging up to 70 µA, and hence establishes itself as a good conductor. On the other hand, the guanine-based device is comparatively less conductive, exhibiting current in the order of 3 µA. Another interesting observation about the guanine-based device is the visibility of a prominent negative differential resistance effect during the positive bias and a tunneling effect during negative bias. The uniform charge transfer through the cytosine device confirms its application as a molecular wire. The observations on the guanine-based device give better insights into its application as a memory device for nano-scale devices.