Volume 14 | Issue 3 |

 Abstract

Abstract: Introduction : Maternal mortality has remained a serious concern for decades, particularly in developing countries. In Niger, despite strategies developed to reduce maternal mortality, progress in the Zinder region remains slow. The objective is to describe the epidemiological aspects and etiological factors of maternal deaths in the Zinder region from 2020 to 2024. Methodology: This was a retrospective descriptive cross-sectional study. All maternal death records from 2020 to 2024 at the Zinder Community Health Center (CSME) were included. Results : A total of 678 maternal deaths were recorded over the five-year period, out of 25,123 admissions and 13,423 live births. The mortality rate was 2.69%, and the maternal mortality ratio (MMR) was 52,051.03 per 100,000 live births. Women under the age of 25 who died represented 45.6% (n=309). 96.02% (n=651) were housewives, and 54% (n=366) died before childbirth. 84.66% (n=574) lived in rural areas, and 51.6% (n=350) had not received prenatal care. 77.28% (n=524) had been referred for medical reasons. Home births accounted for 31.41% of all births. Severe anemia during pregnancy was the primary cause of death in 52.21% of cases. Death occurred in the third trimester in 53.2% of cases. Hemorrhagic causes accounted for 17.7% (n=120), hypertensive disorders such as pre-eclampsia and eclampsia for 18.28% (n=124), and infectious causes for 14.1% (n=96). The majority of deaths, 75.36% (n=511), occurred within the first 24 hours following admission. Conclusion : Maternal mortality in Niger remains a concern, as most deaths occurred antepartum due to preventable causes such as anemia. This is far from achieving the Sustainable Development Goals (SDGs) by 2030.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords : Maternal mortality, CSME, Zinder, etiological factors.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Ananta Nallah, a tributary of Dangori River, acts as a conduit for the Brahmaputra River's water, causing severe river erosion in the region. Since 1992, this erosion has displaced hundreds of families, leading to the loss of their homes and agricultural lands. This has left them homeless and impacted their livelihoods and mental health. The erosion has caused the loss of several lives, leading to emotional and psychological distress in the region. While authorities have deployed temporary measures such as concrete blocks and geo-bags, these have proven ineffective, leading to recurring erosion. The affected residents and local organisations have been actively involved in protesting and demanding effective long-term solutions and rehabilitation from the government. This study aims to explore the socio-cultural impact and displacement due to river erosion in the Kordoiguri, Doom Dooma region in the Tinsukia District, Assam.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Kordoiguri, Brahmaputra River, Socio-cultural impact, Displacement, River Erosion.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Examination anxiety is one of the most common psychological challenges faced by students, particularly at the secondary school level where academic pressure tends to be high. It refers to a set of emotional, cognitive, and physiological responses that arise in evaluative situations such as examinations. While a moderate level of anxiety may act as a motivating force, excessive anxiety can negatively influence students' academic performance, mental health, and overall well-being. The present study aims to examine the level of examination anxiety among secondary school students and to explore its relationship with their academic achievement. A sample of 200 students was selected using a simple random sampling technique to ensure representativeness. Data were collected through a standardized Examination Anxiety Scale and students' academic achievement scores were obtained from school records. The study focused on understanding how anxiety levels vary among students and how these levels impact their performance. The collected data were analyzed using statistical tools such as mean, standard deviation, correlation, and independent samples t-test. The findings revealed that a significant proportion of students experience moderate to high levels of examination anxiety. A strong negative correlation was found between examination anxiety and academic achievement, indicating that students with higher anxiety levels tend to perform poorly in examinations. The study highlights the importance of addressing psychological factors in education and suggests the need for counseling services, stress management programs, and supportive learning environments to help students cope with examination-related stress.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Examination Anxiety, Academic Achievement, Secondary Students, Stress, Psychological Factors

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper presents a RAG-Based Web Intelligence System designed for domain-specific sentiment analysis of financial data. The system integrates Natural Language Processing (NLP) techniques with Retrieval-Augmented Generation (RAG) to analyze real-time web-based information and generate meaningful insights. News data is collected from online sources and processed using a domain-specific sentiment analysis model to classify information into positive, negative, and neutral sentiments. The processed data is transformed into vector embeddings and stored in a vector database for efficient retrieval. The RAG framework enables the system to retrieve relevant contextual information and generate accurate, context-aware responses to user queries. Additionally, the system provides visualization of sentiment trends and integrates stock-related data to support analytical understanding. This approach enhances decision-making by combining data retrieval, machine learning, and generative AI, making it suitable for financial analysis and intelligent web-based applications.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

RAG, Domain Specific Sentiment Analysis, NLP, Financial Data Analysis, Vector Database, ChromaDB, Information Retrieval, Generative AI, Web Intelligence

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Mental health disorders, particularly depression and anxiety, are rapidly increasing due to modern lifestyle stress, social isolation, and lack of timely diagnosis. Early detection remains one of the most critical challenges, as many individuals hesitate to seek professional help due to stigma, lack of awareness, or limited accessibility. EchoSense proposes a hybrid Artificial Intelligence (AI) system that enables early detection of mental health conditions through multimodal data analysis, including voice tone, facial expressions, and textual sentiment. Unlike traditional diagnostic approaches, the system operates in a non-invasive, real-time manner, providing continuous monitoring and personalized feedback. In addition to detection, EchoSense integrates AI-driven music therapy, which dynamically recommends calming, uplifting, or focus-enhancing audio content based on the user's emotional state. The system further enhances user well-being through a mood journal, sleep assistance module, and emergency crisis alert mechanism. In severe cases, automated doctor referral ensures timely professional intervention. The platform emphasizes privacy, scalability, and accessibility, making it suitable for widespread adoption across diverse populations. EchoSense aims to bridge the gap between early diagnosis and continuous emotional support, contributing to preventive mental healthcare and improved quality of life.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence, Mental Health Detection, Multimodal Emotion Recognition, Affective Computing, Natural Language Processing, CNN, LSTM, MFCC, Multimodal Data Fusion, AI-driven Music Therapy, Digital Mental Health, Emotion-Aware Systems, Explainable AI, Predictive Analytics

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Background:Spirituality is increasingly recognized as a vital component of holistic health, influencing how individuals cope with chronic illness and derive meaning from their experiences. However, few validated tools exist that accurately capture spirituality within the Indian cultural context. This study describes the development, validation, and scoring interpretation of the Spirituality Assessment Scale (SAS-20)--a concise, culturally adapted tool designed to assess multiple dimensions of spirituality among individuals with chronic illness. Methods:This single-centre retrospective study was conducted at XXX between October 2021 and March 2022. Tool development followed a two-phase mixed-method approach. Phase I included literature review, focus group discussions, and expert input for item generation and refinement. Phase II involved pilot testing with 10 patients for face validity and evaluation by seven experts for content validation. Psychometric testing assessed internal consistency and content validity. The SAS-20 scoring ranges from 0 to 80, categorized as low, moderate, and high spirituality. Results:The final SAS-20 comprises 20 items across nine dimensions, showing excellent internal consistency (Cronbach's ?=0.858) and strong content validity (CVI=0.8929). Conclusion:The SAS-20 is a brief, reliable, and culturally sensitive tool suitable for assessing spirituality in individuals with chronic illness in clinical and holistic care settings.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Spirituality, Tool validation, Chronic illness, Psychometrics, Holistic health, SAS-20

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Chhattisgarh, often referred to as the "Land of Forests," possesses immense potential for the integrated development of eco-tourism and cultural tourism. With nearly half of its geographical area covered by forests, the state offers a unique combination of natural beauty and rich indigenous heritage. Eco-tourism in the region focuses on sustainable interaction with nature through protected areas such as Kanger Valley National Park and Achanakmar Wildlife Sanctuary, which are known for their biodiversity, scenic landscapes, and conservation initiatives. Natural attractions like Chitrakote Falls further enhance the state's appeal by attracting nature enthusiasts and adventure tourists. Simultaneously, cultural tourism in Chhattisgarh reflects the vibrant traditions of its tribal communities, whose lifestyles, festivals, handicrafts, music, and dance forms represent a living cultural heritage. Historic centers such as Sirpur and the ancient Bhoramdeo Temple showcase the architectural and spiritual legacy of the region. The promotion of traditional crafts, including bell metal and terracotta art, strengthens both cultural preservation and rural livelihoods. The combined growth of eco-tourism and cultural tourism supports environmental conservation, socio-economic development, and the empowerment of local communities. Through responsible tourism practices and sustainable planning, Chhattisgarh holds strong potential to emerge as a model state for balanced and inclusive tourism development in India.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Eco-tourism, Cultural tourism, Community participation, local communities, Heritage Tourism, Sustainable Development, Ecological Integrity, Socio cultural Sustainability.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Urban drainage blockages are a persistent and costly infrastructure challenge across Indian cities, particularly in flood-prone and rapidly urbanising regions. Conventional drainage management relies on periodic manual inspection and reactive clearance, offering no early warning mechanism and consistently failing to prevent flood-level events. This paper presents an AI-Driven Drainage Blockage Prevention System -- a compact, affordable, and autonomous IoT-based monitoring solution that integrates an HC-SR04 Ultrasonic Sensor, an ESP8266 microcontroller, and a buzzer to provide continuous real-time drainage level monitoring with threshold-triggered local alerting. The system operates entirely offline, requires no cloud infrastructure or internet connectivity, and can be deployed at an estimated component cost of Rs. 500-800 per unit. A functional prototype has been assembled, tested across multiple water level conditions, and validated against designed thresholds. The system correctly classifies drainage status as NORMAL, WARNING, and DANGER -- activating the buzzer with appropriate urgency in each case. With a response time consistently within seconds of threshold breach, the proposed system provides an accessible, scalable foundation for intelligent urban drainage management aligned with India's Smart Cities Mission and the United Nations Sustainable Development Goals.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Drainage Monitoring, IoT, Ultrasonic Sensor, ESP8266, Smart Cities, Urban Flood Prevention, Embedded Systems, Real-Time Alerting, Blockage Detection

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Classical ANOVA relies on the assumption of homogeneity of variances, rendering it vulnerable to inflated Type I error rates when heteroscedasticity is present. Although Welch's ANOVA and the Brown-Forsythe test provide adjustments for unequal variances, their relative performance particularly in comparison to permutation-based ANOVA methods under non-normality or small-sample conditions remains insufficiently understood. This study conducted a large-scale simulation to compare four ANOVA methods: the classical F-test, Welch's heteroscedasticity-adjusted test, the Brown-Forsythe test, and permutation ANOVA. The simulations varied group variances, sample sizes (balanced and unbalanced, ranging from 10 to 30 per group), and data distributions (normal, skewed, and heavy-tailed). The classical F-test exhibited inflated Type I error rates under moderate heteroscedasticity. Welch's ANOVA maintained robustness in most scenarios, except in extremely small samples (n < 20). The Brown-Forsythe test showed no consistent advantage over Welch's method, while permutation tests performed well under skewed distributions but experienced reduced power when sample sizes were small. Based on these findings, Welch's ANOVA is recommended for general use due to its balance of robustness and computational efficiency. Permutation tests are preferable in cases of severe skewness or pronounced variance heterogeneity, despite their higher computational cost. These results provide practical guidance for selecting appropriate ANOVA methods based on the specific characteristics of the data.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Heteroscedasticity, Robust ANOVA, Permutation Tests, Type I error, Simulation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Parkinson's disease is defined as Chronic, progressive neurodegenerative disease characterized by both motor and non-motor features. The most important signs and symptoms of Parkinson's disease occur when the nerve cells in basal ganglia, an area of brain that controls the movement become impaired or die. When these neurons die or get impaired, they tend to produce less amount of important chemical in the brain called as Dopamine which causes various symptoms like Tremors, Rigidity in muscles, Akinesia and postural disability associated with various Cognitive, Behavioural and other Psychological symptoms. This disease is increasing in its frequency with the world population showing an incidence of 1-2 per 1000 population and has equal sex distribution. The exact cause of the Parkinson's disease is not exactly known. But the factors such as Genetic predisposition, infections such as Encephalitis, Prolonged use of Antipsychotic drugs, Arteriosclerosis, Neurotoxins etc. can affect the substantia nigra in the basal ganglia leading to the destruction of Dopamine producing neurons. The Vagus nerve has been hypothesized that Alpha-synuclein aggregates form in the enteric nervous system, and spread via the autonomic system to the Central Nervous System. Normally, there is an equilibrium between acetylcholine and dopamine. With dopamine depletion, there results in the hyperactivity of acetylcholine which is the cause of Parkinsonism group of nerve cells deep within the centre of the brain in an area called substantia nigra.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Understanding Parkinsonism through Samprapti of Vatavyadhi

  License

Creative Commons Attribution 4.0 and The Open Definition