SSS. Salaji

S. Salaji

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


S. Salaji has a total of 6 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A reliability-aware deep learning approach for AML subtype classification using dynamic channel–spatial attention

    by Aswathy Elma Aby, Salaji S, Anilkumar K K, Tintu Rajan

    Acute Myeloid Leukaemia (AML) is a haematological malignancy in which abnormal myeloid precursor cells (blasts) multiply uncontrollably in the bone marrow and peripheral blood. It primarily affects...

  • Classification of acute myeloid leukemia by pre-trained deep neural networks: A comparison with different activation functions

    by Aswathy Elma Aby, S. Salaji, K.K. Anilkumar, Tintu Rajan

    Acute Myeloid Leukemia(AML) is a rapidly progressing cancer affecting blood and bone marrow, marked by the swift proliferation of abnormal myeloid cells. Effective treatment requires precise classification of AML subtypes. Conventional classification methods rely on manual microscopic analysis, which is time-consuming and variable, while traditional machine learning approaches often struggle with feature extraction and generalization. This underscores the need for automated detection methods which enhance this process by minimizing manual effort. Activation functions are critical in neural networks, introducing non-linearity that influences training convergence, computational efficiency, and model performance. This study evaluates the effectiveness of CNN architectures, Sequential (VGG16), Directed Acyclic Graph (InceptionV3), and Residual (ResNet50v2) in distinguishing between AML subtypes: AML without maturation, Acute Monoblastic Leukemia, and Pure Erythroid Leukemia, using peripheral blood smear images, while also investigating the impact of different activation functions on model accuracy and training time. The results show that ResNet50v2 achieves the shortest training time, while InceptionV3 takes the longest due to its complex architecture. GELU delivers the highest accuracy, reaching 94.02 % in InceptionV3 and 92.53 % in ResNet50v2, while SELU achieves the highest accuracy for VGG16 at 92.83 %. Mish provides competitive accuracy with lower training time than GELU, while Softplus and Softsign consistently perform poorly. This research demonstrates the potential of CNNs for automating AML subtype classification and identifies GELU as the most effective activation function. Future work could explore data augmentation, optimized activation functions, and attention mechanisms to improve classification performance.


  • Open Access

    Experimental and Molecular Level Analysis of the Tribological and Oxidative Properties of Chaulmoogra Oil

    by S. Salaji, N. H. Jayadas

    This study introduces chaulmoogra oil as a base stock for lubricant formulation. The tribological properties of chaulmoogra oil are evaluated by quantitative structure-property relation (QSPR) technique using the molecular modelling package Spartan 18. The quantum chemical calculations were performed on a typical molecule of chaulmoogra oil and its constituent fatty acids. The orbital energy gap of the constituent fatty acids in chaulmoogra oil is 7.37 eV and that of chaulmoogra oil molecule is 6.8 eV, which is less than that of the lauric acid, the main constituent of coconut oil (7.78 eV). Orbital energy gap predicts a better tribological performance for chaulmoogra oil, and the four ball test result is in agreement with this prediction. Oxidative property of chaulmoogra oil is tested by isothermal thermogravimetric/differential thermal analysis (TGA/DTA) and compared with different oils. Weight gain in oxygen is only 0.02% for chaulmoogra oil and showed better oxidative stability among all other tested oils.


  • Evaluation of physicochemical and tribological properties of chaulmoogra (Hydnocarpus wightianus) oil as green lubricant base stock

    by S Salaji, Nh Jayadas

    Due to environmental concerns, natural oils are increasingly being used as an alternative to mineral oil-based products such as lubricants. However, the use of edible oil for industrial application adversely affects their availability for edible purposes. Chaulmoogra oil is a non-edible oil obtained from the seeds of chaulmoogra tree that grows in wastelands. The fatty acid constituents of vegetable oils determine their properties. Presence of long-chain cyclic fatty acids like chaulmoogric acid, hydnocarpic acid and gorlic acid distinguishes chaulmoogra oil from other oils. Research on how cyclic fatty acids influence the properties of vegetable oils and affect their efficacy as base oils for lubricants is scarce in the literature. This paper presents a preliminary evaluation of the physicochemical, viscometric, oxidative and tribological properties of chaulmoogra oil as a lubricant base stock. Gas chromatography–mass spectroscopy, Fourier transform infrared spectroscopy and nuclear magnetic resonance spectroscopy methods were used for structure elucidation. The viscometric analysis was conducted by using Brookfield viscometer. The tribological performance was evaluated by a four-ball apparatus as per ASTM D4172-94. Thermal and oxidative stabilities were evaluated by thermogravimetric/differential thermal analysis in nitrogen and oxygen environments. Pour point was assessed by ASTM D97 method and differential scanning calorimetry. Chaulmoogra oil has a viscosity of 9.2 cSt at 100 ℃ and a viscosity index of 166. Though the coefficient of friction was better than a commercial lubricant of similar viscosity range (SAE 20W40), the wear scar diameter was comparatively higher. Addition of antiwear additive zinc dialkyl dithiophosphate showed considerable improvement in the wear performance of chaulmoogra oil. Chaulmoogra oil showed a high pour point value around 15 ℃ which needs to be improved by appropriate pour point depressants. Environmental advantage, coupled with non-edible nature of chaulmoogra oil, makes it an excellent base stock for lubricants.