Ananda Chandra College

Ovulation represents a fundamental physiological process in human reproduction, characterized by the release of an oocyte from the ovary. Infertility is a significant reproductive health condition that is frequently associated with ovarian dysfunction. The prevalence of infertility and the utilization of Assisted Reproductive Technologies (ART) has increased globally in recent decades. Trans-abdominal and transvaginal ultrasonography (USG) of the ovaries provide clinically relevant information on follicular count, size, spatial distribution, and their responsiveness to hormonal stimulation. Manual evaluation of large volumes of ultrasound images for follicle identification is both labor-intensive and susceptible to human error. This study proposes a new machine learning architecture, termed the widely integrated follicle extraction network ( WIFE-Net ), which enables efficient and accurate follicle detection from ultrasound images. The model integrates both conventional and “Atrous” convolutional operations to enhance feature extraction from ultrasound (USG) images. Additionally, the model employs two “Pyramid Pooling Operation” to aggregate spatial information across multiple scales, enabling effective representation of object regions from coarse to fine levels. The proposed model facilitates fully automated follicle segmentation with high accuracy. A dataset comprising approximately 64000 annotated 2D ovarian ultrasound images, extracted from UltraSOund Volumes of Annotated ovaries in 3D (USOVA 3D) image volumes, was used to train and evaluate the architecture. Experimental results demonstrate that the proposed model outperforms several state-of-the-art convolutional neural network (CNN)-based approaches on this dataset, highlighting its effectiveness and robustness in clinical applications. Specifically, WIFE-Net achieved an accuracy of 97.78% , Jaccard Similarity Index of 75% , Recall value of 95.84% , F1 Score of 97% , precision of 97% , and Area UnderROC Curve (AUC) of 98% . These results indicate the potential applicability of the proposed method in clinical workflows for reliable follicle assessment.

Urban areas worldwide are witnessing significant population growth and extensive built-up development within their city space and surrounding suburban areas. This results in substantial landscape changes and raises serious concerns regarding the impact on the natural environment. Therefore, monitoring urban dynamics and changes in natural land use is crucial. This study focuses on modeling land-use dynamics and urban growth prediction in a Class-I urban agglomeration in eastern India using remotely sensed data and GISs. The integration of an artificial neural network (ANN) and cellular automata-Markov Chain was used to trace the urban growth patterns and directions for the next two decades. Six key drivers were identified and tested by incorporating Cramer's V statistics. The results indicated that from 2001 to 2021, the built-up area increased by 9.21 km2, while other land-use categories experienced significant declines. Considering the status in 2021, the urban expansion is projected to grow by 18.2% by 2030 and 47.44% by 2040, primarily along road networks and near urban centers. Conversely, vegetation, agricultural land, fallow land, and water bodies are expected to decrease by 9.47%, 28.73%, 4.25%, and 10.11%, respectively, by 2040. These findings provide valuable insights into historical and potential land-use changes and the direction of urban expansion, enabling local governments to develop effective management strategies for sustainable land-use planning and urban growth.

The ecological environment plays a crucial role in maintaining the balance and sustainability of ecosystems, particularly in the context of rapid urbanization. This study investigates the impact of urban growth on ecological environment quality in the Jalpaiguri Planning Area, a rapidly urbanizing area that has been overlooked in previous research. Using remote sensing and geographic information system, this study employed the Remote Sensing Ecological Index (RSEI) to quantitatively assess ecological environment quality by integrating key biophysical properties such as greenness, wetness, dryness, and surface temperature using multi-temporal Landsat data from 1991 to 2021. The results revealed a significant deterioration in eco-environment quality with a decline in mean RSEI values from 0.70 to 0.45. Areas with moderate to excellent ecological quality declined over time, while poor and fair quality zones increased, especially in the urban core and surrounding areas. Moran’s I increased from 0.332 to 0.389, suggesting an increasing spatial dependence and clustering of ecological conditions, indicative of growing environmental polarization. Local indicators of spatial association highlighted a decreasing trend in High-High clusters and an expansion of Low-Low clusters, indicating degradation in greenness and wetness due to intensified built-up development. The outcomes of regression analysis revealed a strong and consistent negative correlation between growing built-up areas and RSEI, with correlation coefficients ranging from −0.76 to −0.86 over the study period. The results support targeted planning interventions, including protection of green and blue spaces, control of unplanned built-up expansion, and integration of RSEI-based ecological monitoring into urban planning for informed decision-making.

The purpose of this study was to investigate whether academic librarians working in government-aided degree colleges in West Bengal, India, possess the competencies necessary to practice embedded librarianship. Employing a mixed-methods design, data were collected through a structured online questionnaire and follow-up telephonic interviews with 116 full-time academic librarians. The study revealed that most librarians demonstrate strong competencies in interpersonal communication, teaching, and collaboration. However, weaknesses were identified in areas such as instructional design, advanced data analysis, and digital content management. Despite their active involvement in embedded librarianship-related activities, more than 80 percent were unaware of the concept prior to the study. Numerous barriers, including staff shortages, heavy workloads, limited institutional support, and lack of recognition, inhibited the broader implementation of embedded librarianship. The findings suggested a need for targeted professional development, increased institutional support, and strategic policy initiatives to enable librarians to fully adopt embedded practices. This was the first empirical study in the Indian context, specifically West Bengal, assessing embedded librarianship readiness among college librarians. It contributes to local and global discourse on transforming academic librarianship.

Abstract This paper examines the influence of physicochemical parameters on drug distribution within vascular tissue following the implantation of a drug-eluting stent (DES) through a mathematical model. The governing equations representing the transport of free drug eluted from DES, and two-phase binding, namely, extracellular matrix (ECM) binding and receptor (REC) binding are solved numerically in an explicit manner. A class of boundary conditions, such as the Dirichlet, Neumann, and Robin types, are used at the mural interface and the perivascular end. Simulations predict that an increase in inter-strut distance amplifies the concentration of all forms but at larger times, this distinction disappears. Concomitantly, a time-dependent release of the drug contributes to the diminishing concentration after attaining its respective peak value. Furthermore, predicted results show a significant difference in the concentrations for no-flux and sink conditions at the interfaces. Moreover, findings show sirolimous drug has a larger mean ECM-bound drug than paclitaxel; however, the trend for REC-bound drug is reversed. Finally, the effect of advection, strut diameter and stent embedment can not be ruled out in this investigation.