Dhanya Dinesh
Dhanya Dinesh has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
- Aleyamma Mathew
Division of Cancer Epidemiology & Biostatistics, Regional Cancer Centre, Thiruvananthapuram, India
2 shared publications - Anila Tresa Alukal
Division of Gynecological Oncology, Regional Cancer Centre, Thiruvananthapuram, India
2 shared publications - Siva Ranjith Jayapalan
Division of Surgical Oncology, Regional Cancer Centre, Thiruvananthapuram, India
2 shared publications - Suchetha Sambasivan
Division of Gynaecologic Oncology, Regional Cancer Centre, Thiruvananthapuram, India
2 shared publications - Ammu Jayanthi Viswanathan
Regional Research Institute of Unani Medicine, Chennai
1 shared publication Anurag Vijay Agrawal
Department of Electronics and Communication Engineering, Bhagwant Institute of Technology, Bhagwantpuram, Muzaffarnagar, Uttar Pradesh 251315, India
1 shared publication
Latest Publications
- 4 September 2025
- Published by Springer Nature in Food Analytical Methods
- Vol. 18 (11) , 2567-2591
Multi-class Fruit Freshness and Adulteration Detection Using Deep Learning Models Optimized by Simulated Annealing and Grad-CAM
- 4 September 2025
- Published by Springer Nature in Food Analytical Methods
- Vol. 18 (11) , 2567-2591
Multi-class Fruit Freshness and Adulteration Detection Using Deep Learning Models Optimized by Simulated Annealing and Grad-CAM
- Open Access
- 31 May 2025
- Published by Science Publishing Corporation in International Journal of Basic and Applied Sciences
- Vol. 14 (2) , 401-409
Deep Learning-Based Classification of Comments and Reviews for Sustainable Development Goals (SDGs) with Web Application Implementation
This work aims to develop an intelligent system that leverages natural language processing (NLP) and deep learning to analyze and interpret textual data in the context Sustainable Development Goals (SDGs). The core objective is to identify semantic relationships between input text and specific SDGs, thereby enabling automated classification and supporting sustainable decision-making. The work focuses on the application of data augmentation techniques to enhance training datasets, refinement of existing classification models through hyperparameter tuning, and the proposal of a novel classification model to improve accuracy and reliability. Additionally, the work includes the development of a user-friendly web application with extended functionalities, allowing users to input text manually or upload text files to determine the most relevant SDG. This integrated approach aims to bridge the gap between unstructured textual data and structured sustainability frameworks, providing an innovative tool for researchers, policymakers, and organizations working toward global sustainability goals.
- 31 May 2025
- Published by Science Publishing Corporation in International Journal of Basic and Applied Sciences
- Vol. 14 (2) , 401-409
Open AccessDeep Learning-Based Classification of Comments and Reviews for Sustainable Development Goals (SDGs) with Web Application Implementation
This work aims to develop an intelligent system that leverages natural language processing (NLP) and deep learning to analyze and interpret textual data in the context Sustainable Development Goals (SDGs). The core objective is to identify semantic relationships between input text and specific SDGs, thereby enabling automated classification and supporting sustainable decision-making. The work focuses on the application of data augmentation techniques to enhance training datasets, refinement of existing classification models through hyperparameter tuning, and the proposal of a novel classification model to improve accuracy and reliability. Additionally, the work includes the development of a user-friendly web application with extended functionalities, allowing users to input text manually or upload text files to determine the most relevant SDG. This integrated approach aims to bridge the gap between unstructured textual data and structured sustainability frameworks, providing an innovative tool for researchers, policymakers, and organizations working toward global sustainability goals.
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3193 (1) , 020137
Improving personalized predictions through nonlinear individual-specific parameters in neural network-based mixed effect models
For sequential tasks, such as daily mood ratings, personalized prediction—a machine learning method— can be useful since it predicts future observations based o For sequential tasks, such as daily mood ratings, personalized prediction—a machine learning method— can be useful since it predicts future observations based o
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3193 (1) , 020137
Improving personalized predictions through nonlinear individual-specific parameters in neural network-based mixed effect models
For sequential tasks, such as daily mood ratings, personalized prediction—a machine learning method— can be useful since it predicts future observations based o For sequential tasks, such as daily mood ratings, personalized prediction—a machine learning method— can be useful since it predicts future observations based o
- 3 March 2022
- Published by Springer Nature in Indian Journal of Surgical Oncology
- Vol. 13 (3) , 580-586
Clinical Profile and Survival Outcome of Endometrial Cancer with p53 Mutation
Clinicopathologic classification of endometrial cancer imperfectly reflects the tumor biology. Pathologic categorization — especially in high-grade tumors — results in an imprecise estimation of the risk of disease, recurrence, and death. Molecular subtyping is emerging as the standard of care in diagnosis and treatment of endometrial cancers. Molecular markers are important prognostic factors in tumor dissemination and early recurrence of endometrial cancers. TP53 mutation is an important prognostic factor for both serous and endometrioid cancers. The study aims to compare the clinical profile and overall survival of endometrial cancers with and without p53 mutation. Sixty-three patients who underwent surgical staging for carcinoma endometrium were included in the study.TP53 mutation status was determined based on p53 expression by immunohistochemistry (IHC) as a p53 wild or p53 mutant type. Data were analyzed for the clinical profile, p53 mutation status on IHC, histological pattern, tumor grade, stage of the disease, lymph node spread, recurrence pattern, treatment received, 2-year disease-free survival, and overall survival. Recurrence was noted in 12.7% patients after 2-year follow-up, of which 75% patients had p53 mutation. Significant association was seen between p53 expression and high-grade tumors, stage, cervical involvement, and adnexal involvement. The 2-year overall survival of the p53 wild type was 97.2% and the p53 mutant type was 91.7%. The 2-year disease-free survival for the p53 wild type was 94.3% and the disease-free survival of the p53 mutant variety was 83.5%. The 2-year disease-free survival for endometrioid carcinoma with p53 wild type was 100% and p53 mutant variety was 86.2% (p value 0.033). About 15.9% (10) patients were reassigned to the high-risk group needing chemotherapy and radiation according to the ESGO ESTRO 2021 consensus classification, due to their p53 mutation status. IHC to assess somatic p53 mutation may be done in endometrial biopsies irrespective of their histology. This may help to identify that the aggressive tumors thereby help in tailoring surgery, planning adjuvant treatment, and follow-up.
- 3 March 2022
- Published by Springer Nature in Indian Journal of Surgical Oncology
- Vol. 13 (3) , 580-586
Clinical Profile and Survival Outcome of Endometrial Cancer with p53 Mutation
Clinicopathologic classification of endometrial cancer imperfectly reflects the tumor biology. Pathologic categorization — especially in high-grade tumors — results in an imprecise estimation of the risk of disease, recurrence, and death. Molecular subtyping is emerging as the standard of care in diagnosis and treatment of endometrial cancers. Molecular markers are important prognostic factors in tumor dissemination and early recurrence of endometrial cancers. TP53 mutation is an important prognostic factor for both serous and endometrioid cancers. The study aims to compare the clinical profile and overall survival of endometrial cancers with and without p53 mutation. Sixty-three patients who underwent surgical staging for carcinoma endometrium were included in the study.TP53 mutation status was determined based on p53 expression by immunohistochemistry (IHC) as a p53 wild or p53 mutant type. Data were analyzed for the clinical profile, p53 mutation status on IHC, histological pattern, tumor grade, stage of the disease, lymph node spread, recurrence pattern, treatment received, 2-year disease-free survival, and overall survival. Recurrence was noted in 12.7% patients after 2-year follow-up, of which 75% patients had p53 mutation. Significant association was seen between p53 expression and high-grade tumors, stage, cervical involvement, and adnexal involvement. The 2-year overall survival of the p53 wild type was 97.2% and the p53 mutant type was 91.7%. The 2-year disease-free survival for the p53 wild type was 94.3% and the disease-free survival of the p53 mutant variety was 83.5%. The 2-year disease-free survival for endometrioid carcinoma with p53 wild type was 100% and p53 mutant variety was 86.2% (p value 0.033). About 15.9% (10) patients were reassigned to the high-risk group needing chemotherapy and radiation according to the ESGO ESTRO 2021 consensus classification, due to their p53 mutation status. IHC to assess somatic p53 mutation may be done in endometrial biopsies irrespective of their histology. This may help to identify that the aggressive tumors thereby help in tailoring surgery, planning adjuvant treatment, and follow-up.
- 25 July 2019
- Published by Springer Nature in Peer-to-Peer Networking and Applications
- Vol. 12 (5) , 1194-1213
Dolphin partner optimization based secure and qualified virtual machine for resource allocation with streamline security analysis
- 25 July 2019
- Published by Springer Nature in Peer-to-Peer Networking and Applications
- Vol. 12 (5) , 1194-1213
Dolphin partner optimization based secure and qualified virtual machine for resource allocation with streamline security analysis