Extended Abstract
Background and Purpose
The aging population has become a serious problem, particularly in developing countries. The increase in the elderly population and the associated costs are among the most significant socio-economic and health challenges worldwide. Numerous studies in public health, urban planning, and environmental psychology have demonstrated the relationship between the physical and mental health of older adults and various aspects of the physical and social environment—specifically as this relationship pertains to the health of older adults through physical activity. Despite the significant increase in research on physical activity in recent years, no scientometric study has been conducted on this approach at the housing level, especially in neighborhood spaces. Accordingly, the present study seeks to systematically evaluate and analyze these research outputs and, in doing so, take a step toward clarifying the scientific status of this field and charting its future directions.
Methods
To collect data in this stage, a qualitative research approach, content analysis method, and systematic review with co-word analysis technique were employed. To identify high-frequency keywords, focal areas, and low-citation research fronts within this domain, a comprehensive search strategy was implemented. Scientific publications were selected in consultation with experts in architecture specializing in housing, using Boolean operators 'AND' and 'OR' in the document search section. The search was conducted across article titles, abstracts, and keywords in the Scopus electronic database, and the output was exported in CSV format. The framework and principles of this systematic review adhered to the PRISMA checklist. After applying the predefined inclusion and exclusion criteria and conducting a re-review, 176 full-text sources were included in this systematic review. The author keywords from all extracted sources were organized into a co-occurrence matrix within an Excel file and subsequently transferred to VOSviewer software (version 18.6.1). To analyze keyword co-occurrence, a threshold of two was set experimentally for author keywords in the software input. Following the removal of irrelevant terms, the remaining 83 keywords were mapped and analyzed using the co-occurrence network in the software. The findings identified 12 clusters related to the topic of older adults' physical activity in neighborhood spaces. Following the scientometric analysis (Figure 2), the most cited journals, leading countries, key variables, software, methods, and tools used in scientific productions within this field were identified according to the presented definitions.
Results
To identify high-frequency and low-frequency keywords in the field of older adults' physical activity in neighborhood spaces, the co-word analysis method was employed. VOSviewer software identified 12 clusters by analyzing the keywords of the abstracts from 176 sources included in the systematic review. The first and second clusters, shown in red and green, consist of 14 and 10 keywords, respectively. In the first cluster, the two keywords 'neighborhood environment' and 'physical environment,' with link strengths of 25 and 22 respectively, ranked among the most frequent keywords. In the second cluster, 'elderly' and 'GPS,' with total link strengths of 28 and 25, were the most frequent keywords.
Scientific production in this field followed an upward trend from 2000 to 2016 and a downward trend from 2017 to 2019. In 2020 and 2021, production resumed an upward trajectory, followed by a significant decline from 2021 onward, as illustrated in Figure 4. The countries that have contributed to this research domain were also examined and analyzed. The leading positions in scientific production belonged to the United States and China, with 34 and 32 sources, respectively. Canada and Japan ranked third and fourth, with 15 and 14 scientific productions.
Among the methods used in the studies included in the systematic analysis, descriptive and correlational methods were the most frequently employed, with frequencies of 161 and 135, respectively. Following these, the survey-cross-sectional method ranked third. The pretest-posttest method ranked lowest, being used in only two articles. The findings indicated that the use of questionnaires, employed in 114 scientific publications, ranked highest among the tools for measuring older adults' physical activity in residential settings. The interview technique was the second most frequent tool, with a frequency of 93.
The variables density, green space, aesthetics, and street connectivity were among the most frequent variables, with frequencies of 98, 97, 67, 65, and 55, respectively. The variables geometry, complexity, scalability, and orientation ranked lowest, each with a frequency of one. SPSS software was the most widely used software, employed in 57 scientific productions. Stata software ranked second, with a frequency of 28. The lowest ranks belonged to MPlus and AtlasTi software, each with a frequency of two.

Conclusion
The results of this study indicate a significant research gap in studies related to the application of artificial intelligence, physical activity measurement tools, and built environment components within this domain. The findings underscore the necessity of expanding interdisciplinary research and strengthening scientific collaborations between Iran and both developed and developing countries in future studies. It is noteworthy that relying on a single database presents a limitation; focusing exclusively on Scopus may have reduced the scope of data coverage. Therefore, employing additional databases such as Web of Science and PubMed is recommended for future research.
Overall, the findings of this review can serve as a guide for researchers in identifying priority research areas, reputable journals, and prominent authors for publishing articles or developing scientific collaborations. Furthermore, these results provide a more comprehensive understanding of the position of physical activity among older adults in the context of residential open spaces by offering a clearer articulation of research trends and highlighting emerging research issues.

Article Message
The results of this study highlight a substantial research gap in the domain of artificial intelligence, physical activity measurement tools, and the built environment within this field. By offering a clearer articulation of research trends and identifying contemporary hot topics, these findings provide a comprehensive picture of the status of physical activity among older adults in the context of residential open spaces
Ethical Considerations
All research ethical guidelines were observed. The study received approval from the Department of Architecture, Faculty of Engineering, University of Islamic Azad University, Pardis Branch.
Authors' Contributions
Conceptualization: Zahra Eivazi Damirchi, Amir Sanaee & Maryam Cheshmeh Ghsabani
Data Collection: Zahra Eivazi Damirchi
Data Analysis: Zahra Eivazi Damirchi
Manuscript Writing: Zahra Eivazi Damirchi
Review and Editing: Amir Sanaee & Maryam Cheshmeh Ghsabani
Responsible for funding: Zahra Eivazi Damirchi, Amir Sanaee & Maryam Cheshmeh Ghsabani
Literature Review: Zahra Eivazi Damirchi
Project Manager: Amir Sanaee
Conflict of Interest
The authors declare no conflicts of interest.
Acknowledgments
This article is an excerpt from Zahra Eyvazi Damirchi's doctoral thesis titled "Developing an Architectural Model of Neighborhood Spaces in Residential Complexes to Increase Pedestrianization for the Elderly," which is being conducted under the guidance of Dr. Amir Sana'i and the advice of Dr. Maryam Cheshmeh Qasaban at the Faculty of Engineering, Pardis Branch, Islamic Azad University