February 02, 2026
Data intelligence in postoperative care for lipedema: applied visualization and decision support
Caio Santos Ferarezi; Fernando Celso de Campos
Summary prepared by the ResumeAI tool, an artificial intelligence solution developed by Instituto Pecege focused on synthesis and writing.
The objective of this work is to develop an interactive managerial dashboard to consolidate and visualize historical data of post-operative care for patients with lipedema, highlighting trends and patterns to provide strategic insights for the management of a company in the segment. The research aims to transform fragmented operational data into a strategic asset, enabling performance monitoring, anticipation of seasonal demands, and identification of optimization opportunities. The application of business intelligence seeks to fill a managerial gap; the absence of a consolidated view of the data limited planning, replacing manual processes with a data-driven culture.
The research context is the growth of the plastic surgery market in Brazil, one of the global leaders in procedures (ISAPS, 2022). In this scenario, surgery for the treatment of lipedema gains prominence. Lipedema is a chronic condition of disproportionate adipose tissue accumulation, mainly in the lower limbs, which predominantly affects women, causing pain, functional limitation, and impact on quality of life (Kruppa et al., 2020). Frequently confused with obesity, its treatment may involve tumescent liposuction, which requires specialized postoperative follow-up to ensure recovery and results (Forner-Cordero et al., 2021).
The growing demand for this procedure drives the demand for post-operative rehabilitation services, creating a competitive market niche. For companies in the sector, the ability to understand trends, optimize the allocation of professionals, and manage the volume of services is a competitive advantage. Data-driven management becomes fundamental, allowing decisions to be based on evidence extracted from operations. The systematic use of data to guide strategy increases the predictability of results and strengthens business competitiveness (Davenport, 2006).
Data-driven management is materialized in managerial panels, or dashboards, visualization tools that present key performance indicators (KPIs) clearly and interactively. These panels translate large volumes of data into visual representations, such as charts and tables, which facilitate the monitoring of metrics, the identification of trends, and the detection of anomalies (Kumar & Belwal, 2017). By centralizing critical information, dashboards optimize analysis time and empower managers to understand the historical behavior of the business, increasing decision-making assertiveness.
Among the Business Intelligence (BI) platforms, Microsoft Power BI stands out for its robustness, integration with data sources, and ability to promote “data storytelling”, the construction of narratives from data (Becker & Gould, 2019). The tool allows the creation of interactive visualizations, enabling users to explore data from different perspectives. The combination of data, visual elements, and narrative enhances the generation of actionable insights and aligns analytical results with organizational goals, democratizing access to information (Echeverria et al., 2021; Town & Thabtah, 2019).
The methodological approach was structured in four stages to transform raw data into a strategic management dashboard. The research, of an applied and exploratory-descriptive nature, used a historical dataset from a specialized company, with financial records of transfers to employees as the primary source. The process involved extraction, transformation, and loading (ETL) techniques, dimensional modeling, and data visualization to support decision-making.
The first stage consisted of requirements gathering and process mapping to align the solution with business needs. Interviews with managers and stakeholders helped to understand workflows and identify the most relevant metrics and informational bottlenecks, such as the lack of standardization and data decentralization. This mapping, according to Harmon (2019), was essential to define the project scope and ensure that the dashboard answered critical business questions.
The second and third stages, integrated, covered technical understanding, data extraction, and treatment. Initially, Power Query was evaluated to extract data from hundreds of PDF files, but the tool presented limitations in automating recursive reading. The Python language was chosen, offering greater flexibility and control. A script was developed to read the files, extract fields such as patient name, date, type of service, and values, and consolidate everything into a tabular database. Subsequently, a rigorous treatment process was applied, including data cleaning, handling of missing and duplicate values to ensure consistency (Redman, 2001), variable standardization, and creation of aggregations. A fundamental step was the anonymization of sensitive patient and employee data, replacing names with unique identifiers to protect privacy.
The fourth stage was the development of the management dashboard in Microsoft Power BI. With the treated database, dimensional modeling (Kimball & Ross, 2013) was initiated, creating a fact table (tbatendimentos) and a calendar dimension (dcalendario) for temporal analysis. The dashboard design followed best practices in data visualization, prioritizing clarity and objectivity (Few, 2004; Tufte, 2001). Interactive charts, tables, and KPIs were implemented. Analytical measures, such as Year-over-Year growth calculations, were developed using DAX (Data Analysis Expressions) language, essential for temporal intelligence (Ferrari & Russo, 2019). Interactivity was ensured by filters and slicers, allowing for flexible information exploration (Shneiderman, 1996). The dashboard was validated with stakeholders to ensure its adherence to needs and usability, considering that aesthetic perception influences tool acceptance (Cawthon & Moere, 2007).
The results obtained with the management dashboard in Power BI revealed valuable strategic insights. The dashboard was designed in a single-page view, organized into three sections that offer a holistic view of operational and financial performance. Interactivity allows managers to filter data by multiple criteria, transforming the dashboard into a dynamic exploration tool. The consolidation of data, previously dispersed, into a single visual interface represents the first tangible result of the project.
The first section, “Big Numbers”, presents the most important aggregated indicators through visual cards. This section offers a snapshot of the operation, consolidating data from January 2021 to April 2025. The results show 30 active female employees, 3,363 distinct patients, and 26,042 consultations. Financially, the total transfer to professionals amounted to R$ 2.33 million, with an average value per consultation of R$ 89.00. These numbers quantify the company’s volume and serve as a baseline for performance monitoring and goal setting.
The second section, Historical Analysis, uses a line graph to visualize the evolution of the volume of services and financial transfers, with a “drill-down” functionality to change granularity (month, quarter, year). The analysis revealed a clear seasonality pattern: a peak in activity in the third quarter (July to September), with growths such as 79.8% in the third quarter of 2022 compared to the second. In contrast, the fourth quarter (October to December) shows a drop in demand, such as a reduction of 31.2% in the fourth quarter of 2024. This insight suggests that demand is tied to periods of greater surgical procedures, allowing the company to plan personnel allocation, cash management, and marketing actions.
The third section focuses on Employee and Services Analysis. A horizontal bar chart highlighted a strong concentration of revenue in a few professionals. The top five employees were responsible for over 40% of the total amount paid, with “Employee 011” standing out, accumulating R$ 305 thousand in transfers. This finding points to a dependence on key professionals, which represents both a recognition of performance and an operational risk to be managed with retention and development programs.
The analysis by service type revealed an even greater concentration. The “Lipedema Drainage” procedure accounted for over R$ 1.3 million in reimbursements, followed by “Drainage + Taping”, with R$ 377,000. Together, they represented over 70% of the revenue. This data confirms the company’s specialization but exposes a vulnerability associated with low portfolio diversification. Management can use this information to explore complementary services or reinforce its leadership in current services.
To deepen the performance analysis, a dynamic scatter plot crosses metrics such as number of consultations, patients, and referrals. Average lines divide the graph into four performance quadrants, facilitating the identification of profiles. The analysis revealed that physiotherapists stood out for higher referrals, associated with a greater variety of consultations, while nurses, despite attending many patients, registered a lower average referral value. This segmentation allows management to develop targeted training and remuneration strategies.
The discussion of the results evidences the project’s success in transforming raw data into a high-impact analytical tool. The Power BI dashboard centralized and made the information actionable. The identification of seasonality patterns provides an empirical basis for resource planning, replacing intuition with a proactive approach. The quantification of revenue concentration in a few professionals and services illuminates strategic risks that can be mitigated. The positive feedback from the partner company validates the relevance of the solution.
Despite the advances, the study has limitations. The analysis was based on data from a single company, which prevents the generalization of the results. The quality of the data, dependent on manual records, may contain inaccuracies. The study did not incorporate external variables, such as the macroeconomic context, nor quality metrics, such as patient satisfaction. The continued use of the tool will depend on the engagement of the internal team to perform periodic updates.
The present study demonstrated how the application of data analysis and visualization techniques with Microsoft Power BI can generate value for the management of a company in the healthcare sector. The solution enabled the structuring of historical data, making it possible to identify patterns such as demand seasonality, remittance concentration, and performance profiles. These findings provide subsidies for resource optimization, strategic planning, and the promotion of a data-driven culture, reinforcing the potential of business intelligence to increase efficiency in the sector.
The study’s limitations open avenues for future research. Expanding the database to include multiple companies is recommended, allowing for comparative analyses. The integration of clinical indicators and satisfaction metrics would provide a more complete view of performance. Additionally, the application of predictive modeling techniques could be explored to forecast future demands with greater accuracy. It is concluded that the objective was achieved: it was demonstrated that the development of an interactive managerial dashboard is an effective tool for extracting strategic insights from historical patient care data, supporting the management of a lipedema post-operative company.
References:
Becker, L. T.; Gould, E. M. 2019. Microsoft Power BI: Extending Excel to manipulate, analyze, and visualize diverse data. Serials Review.
Cawthon, N.; Moere, A. 2007. The effect of aesthetic on the usability of data visualization. IEEE Transactions on Visualization and Computer Graphics 13(6): 1121–1128.
Davenport, T. H. 2006. Competing on analytics. Harvard Business Review 84(1): 98–107.
Echeverria, V.; Lawrence, L.; Tsai, Y. S.; Singh, S.; Martinez-Maldonado, R.; Fernandez-Nieto, G. 2021. A tutorial on data storytelling techniques for learning analytics dashboards. Companion Proceedings of the 11th International Conference on Learning Analytics & Knowledge.
Few, S. 2004. Show Me the Numbers: Designing Tables and Graphs to Enlighten. Analytics Press, Burlingame, CA, USA.
Forner-Cordero, I.; Forner-Cordero, A.; Szolnoky, G. 2021. Update in the management of lipedema. International Angiology 40(0): 1–11.
Harmon, P. 2019. Business Process Change: A Business Process Management Guide for Managers and Process Professionals. 4ed. Morgan Kaufmann, Cambridge, MA, USA.
International Society of Aesthetic Plastic Surgery [ISAPS]. 2022. International Study on Aesthetic/Cosmetic Procedures Performed in 2022.
Kimball, R.; Ross, M. 2013. The Data Warehouse Toolkit: The Definitive Guide to Dimensional Modeling. 3ed. John Wiley & Sons, Indianapolis, IN, USA.
Kruppa, P.; Georgiou, I.; Biermann, N.; Prantl, L.; Klein-Weigel, P.; Ghods, M. 2020. Lipedema—pathogenesis, diagnosis and treatment options. Deutsches Ärzteblatt International 117(25): 396–403.
Microsoft. 2025. Power BI Documentation.
Paolacci, S.; Precone, V.; Acquaviva, F.; Chiurazzi, P.; Fulcheri, E.; Pinelli, M.; Buffelli, F.; Michelini, S.; Herbst, K. L.; Unfer, V.; Bertelli, M. 2019. Genetics of lipedema: new perspectives on genetic research and molecular diagnoses. European Review for Medical and Pharmacological Sciences 23(12): 5581–5594.
Redman, T. C. 2001. Data Quality: The Field Guide. Digital Press, Burlington, MA, USA.
Russo, M.; Ferrari, A. 2019. Chapter 8 – Time intelligence calculations. In: The Definitive Guide to DAX: Business Intelligence for Microsoft Power BI, SQL Server Analysis Services, and Excel. Microsoft Press, Redmond, WA, USA.
Shneiderman, B. 1996. The eyes have it: a task by data type taxonomy for information visualization. Proceedings of the IEEE Symposium on Visual Languages 5(3): 336–343.
Town, P.; Thabtah, F. 2019. Data analytics tools: A user perspective. Journal of Information & Knowledge Management 18(1): 1950002.
Tufte, E. R. 2001. The Visual Display of Quantitative Information. 2ed. Graphics Press, Cheshire, CT, USA.
Executive summary from the Final Course Work of the Specialization in Data Science and Analytics of the MBA USP/Esalq
Learn more about the course; click here: