The health-related complications such as diabetes, macular degeneration, inflammatory conditions, ageing and fungal infections may cause damages to the retina and the macula of the eye, leading to permanent vision loss. The major diseases associated with retina are Arteriosclerotic retinopathy (AR), Central retinal vein occlusion (CRVO), Branch retinal artery occlusion (BRAO), Coat's disease (CD) and Hemi-Central Retinal Vein Occlusion (HRVO). The symptomatic variations among these disorders are relatively confusing so that a systematic diagnostic strategy is difficult to set in. Therefore, an early detection device is required that is capable of differentiating the various ophthalmic complications and thereby helping in providing the right treatment to the patient at the right time. In this research work, 'Deep Convolution Neural Networks (Deep CNN) based machine learning approach has been used for the detection of the twelve major retinal complications from the minimal set of fundus images. The model was further cross-validated with real-time fundus images. The model is found to be superior in its efficiency, specificity and ability to minimize the misclassification. The “multi-class retinal disease” model on further cross-validation with real-time fundus image of the gave an accuracy of 95.63 %, validation accuracy of 92.99 % and F1 score of 91.96 %. The multi-class model is found to be a theranostic clinical support system for the ophthalmologist for diagnosing different kinds of retinal problems, especially BRAO, BRVO, CRAO, CD, DR, HRVO, HP, HR, and CN.

As image processing techniques constantly grow in complexity & volume, meeting the required demand for data storage and computational power is a challenge. Using hybrid quantum-mechanical systems to encode and process image information could help overcome such challenges. We propose to implement a hybrid quantum mechanical system with 2 qubits operation for the purpose of classifying between cancerous and non-cancerous pigmented skin-lesions in the HAM10000 dataset. In view of the ever-increasing number of skin cancer deaths, such a system could have potential for the early diagnosis of the disease. Until fully scalable quantum hardware becomes available, the hybrid model could be a viable alternative to overcome the limitations of classical computing systems.

As breast cancer remains leading cause of cancer death globally, it is essential to develop an affordable breast cancer therapy in underdeveloped countries. Drug repurposing offers potential to address gaps in breast cancer treatment. Molecular networking studies were performed for drug repurposing approach by using heterogeneous data. The PPI networks were built to select the target genes from the EGFR overexpression signaling pathway and its associated family members. The selected genes EGFR, ErbB2, ErbB4 and ErbB3 were allowed to interact with 2637 drugs, leads to PDI network construction of 78, 61, 15 and 19 drugs, respectively. As drugs approved for treating non cancer-related diseases or disorders are clinically safe, effective, and affordable, these drugs were given considerable attention. Calcitriol had shown significant binding affinities with all four receptors than standard neratinib. The RMSD, RMSF, and H-bond analysis of protein-ligand complexes from molecular dynamics simulation (100 ns), confirmed the stable binding of calcitriol with ErbB2 and EGFR receptors. In addition, MMGBSA and MMP BSA also affirmed the docking results. These in-silico results were validated with in-vitro cytotoxicity studies in SK-BR-3 and Vero cells. The IC50 value of calcitriol (43.07 mg/ml) was found to be lower than neratinib (61.50 mg/ml) in SK-BR-3 cells. In Vero cells the IC50 value of calcitriol (431.05 mg/ml) was higher than neratinib (404.95 mg/ml). It demonstrates that calcitriol suggestively downregulated the SK-BR-3 cell viability in a dose-dependent manner. These implications revealed calcitriol has shown better cytotoxicity and decreased the proliferation rate of breast cancer cells than neratinib. Communicated by Ramaswamy H. Sarma

The recently introduced deep learning environment has been used in post pharmacogenomic predictions on proneness to lung cancer. The well-established pre-trained model, VGG19 has been used to extract information from the pathological images specific to PTEN, EGFR, ERBB2, BRAF and CDKN2A for Lung cancer. The model finds to be suitable for the prediction of these mutations, specific to Indian populations. A model consisting of testing of specific genetic signatures and the deep convolution network based image classification has been suggested as an ingenious technique for the fast, cheap and effective predictions on mutations of lung cancer among Indian populations.

Lung cancer is the leading cause of cancer-related fatalities globally, accounting for the highest mortality rate among both men and women. Mutations in the epidermal growth factor receptor (EGFR) gene are frequently found in non-small cell lung cancer (NSCLC). Since curcumin and CB[2]UN support various medicinal applications in drug delivery and design, we investigated the effect of curcumin and CB[2]UN-based drugs in controlling EGFR-mutant NSCLC through a dodecagonal computational approach. Molecular docking studies revealed that the ligands curcumin (-6.9 kcal/mol) and CB[2]UN (-8.1 kcal/mol) bound more strongly to the EGFR-mutant NSCLC proteins with 2ITX and 2ITV, respectively. Molecular dynamics simulation (50 ns) investigation of protein-ligand complexes using RMSD, RMSF, Rg, and SASA indicated that curcumin and CB[2]UN with EGFR-mutant proteins are kinetically stable. In addition, MMPBSA/MMGBSA analysis confirmed the thermodynamic stability of each curcumin and CB[2]UN protein-ligand complex. Finally, KDeep absolute binding affinity calculations show energies of -6.13 kcal/mol and − 5.26 kcal/mol for 2ITX-CUR and 2ITV-CB[2]UN protein-ligand complexes, respectively. Thus, our dodecagonal strategy reveals that 2ITX-CUR and 2ITV-CB[2]UN are more likely to form protein-ligand complexes with more significant binding affinities and excellent stability throughout the 50 ns simulation time.