Interpretable Decision Tree Model for Classification of Acute Appendicitis Using Clinical, Laboratory, and Imaging Data: A Retrospective Single-Center Study.

By /
Cite this article Download PDF DOI Email Share
Ali Bekraki, Ahmad Shahin

Volume 9 · Issue 2 · pp. 506–514

Published: 2026-06-30

Abstract

Background: Acute appendicitis remains a common surgical emergency, and diagnostic uncertainty persists despite the use of clinical scores and imaging. We examined whether an interpretable decision tree could support classification of appendicitis using routinely available clinical, laboratory, and imaging variables. Methods: This retrospective single-center study included 281 patients who presented to El-Youssef Hospital with suspected appendicitis between June 2015 and May 2020. A J48 decision tree was developed in WEKA using age, fever, white blood cell count, neutrophil percentage, C-reactive protein, ultrasound findings, and computed tomography findings. Model performance was estimated with 10-fold cross-validation and reported using the confusion matrix. Results: The decision tree achieved an overall accuracy of 84.3%, with a sensitivity of 89.8%, a specificity of 76.3%, a positive predictive value of 84.8%, and a negative predictive value of 83.7%. Ultrasound, computed tomography, white blood cell count, and neutrophil percentage were the most informative predictors. Misclassifications were most common in patients with atypical presentations or borderline laboratory values. Conclusion: In this single-center retrospective cohort, a decision tree classifier showed good internal performance for distinguishing appendicitis from non-appendicitis. The findings support the feasibility of an interpretable decision-support model, but they do not establish clinical effectiveness. External multicenter validation and comparison with alternative algorithms and existing scoring systems are needed before clinical implementation.

Keywords: Acute appendicitis; data mining; decision tree; machine learning; diagnostic classification; clinical decision support

Introduction

Improving diagnostic accuracy while avoiding unnecessary intervention remains an important challenge in emergency surgical care1. Acute appendicitis accounts for a substantial proportion of emergency surgical admissions, and delays in diagnosis may increase the risk of perforation, peritonitis, and prolonged hospitalization2,3. Appendectomy remains the definitive treatment and one of the most frequently performed emergency abdominal surgeries. At the same time, unnecessary appendectomy remains clinically relevant because avoidable surgery contributes to healthcare costs and procedure-related morbidity4.

Multiple clinical scoring systems, including the Alvarado, RIPASA, and Tzanakis scores, have been introduced to assist diagnostic stratification in suspected appendicitis2,3,5-7. Even so, diagnostic performance remains variable across settings and patient groups. Because diagnostic performance varies across patient populations and clinical settings, interest has grown in machine-learning approaches capable of integrating multiple diagnostic variables simultaneously.

Advances in computational analysis have expanded the use of data-driven models in clinical decision support research. These approaches aim to identify relationships within complex datasets that may not be readily apparent using conventional analytical methods8. Initially applied in healthcare in the early 1990s, data mining has gained traction in recent years due to advancements in information technology and the widespread availability of electronic medical records9. In surgical practice, such models may help organize heterogeneous diagnostic information into reproducible classification frameworks.

The present study investigated whether a transparent decision tree classifier could distinguish appendicitis from non-appendicitis using routinely collected emergency department data. A decision tree algorithm was selected because its rule-based structure permits direct visualization of how individual predictors contribute to classification outcomes.

Table 1. Attributes and their descriptions that are significant to diagnose acute appendicitis

Table 1

Attribute Description
Nausea and Vomiting The patient went to the emergency department (ER) with this symptom.
Abdominal pain The patient went to the ER with this symptom.
Tenderness The doctor localizes the presence or absence of right lower quadrant tenderness.
Age A numeric value between 0 and 100
Sex Male or Female
Fever Numerical values ranging from less than 37°C to more than 40°C
White Blood Cell (WBC) Numerical values ranging from less than 10,000 Cells/mm³ to more than 20,000 Cells/mm³
Neutrophilia Numerical values ranging from less than 70% to more than 70%
C Reactive Protein (CRP) Numerical values ranging from less than 5 to more than 5
Urine The presence of significant quantity of Red Blood Cell (RBC) or WBC in the urine
Lymphocyte Numerical values ranging from less than 50% to more than 50%
Monocyte Numerical values ranging from less than 10% to more than 10%
Eosinophil Numerical values ranging from less than 5% to more than 5%
Ultrasound (US) Is a string attribute
Computerized Tomography (CT) Is a string attribute
Appendectomy Yes or No

Materials and Methods

2.1.Study Design and Data Collection

This retrospective single-center study included patients who presented to the emergency department of El-Youssef Hospital in Akkar, Lebanon, with suspected acute appendicitis between June 2015 and May 2020. The study aimed to evaluate the performance of an interpretable decision tree classifier for distinguishing appendicitis from non-appendicitis using routinely available clinical, laboratory, and imaging variables.

Patients were eligible if they had a provisional diagnosis of acute appendicitis and a complete medical record, including a final histopathological report for the resected appendix. Patients with incomplete records were excluded. A total of 281 patients were included in the final analysis. Reporting of the study was guided by TRIPOD principles for prediction model research.

The extracted variables were age, sex, fever, white blood cell count, neutrophil percentage, C-reactive protein, urinalysis results, ultrasound findings, computed tomography findings, and appendectomy status. The reference standard outcome was final histopathology, classified as appendicitis or non-appendicitis. The complete list of attributes is presented in Table 1.

2.2.Ethical Approval

This study was conducted in accordance with the principles of the Declaration of Helsinki. Ethical approval for the study was obtained from the Scientific Committee of the Lebanese University, Faculty of Economics and Business Administration, 3rd Branch (approval date: April 2015; approval number: 12022015). Due to the retrospective design of the study and the use of fully anonymized patient data, the requirement for individual informed consent was waived by the ethics committee. Hospital administrative permission was obtained prior to data collection, and strict measures were implemented to ensure patient confidentiality throughout the study.

2.3.Data Pre-Processing and Feature Selection

Data cleaning, integration, and transformation were performed before model development. Variables containing substantial missing or inconsistent data were excluded before model construction. Symptoms such as nausea and vomiting, and urinalysis results, were also removed because they demonstrated little discriminatory value in this dataset. Continuous variables were categorized to facilitate decision tree construction; however, this transformation was used for model building rather than as a claim of clinically validated thresholds.

2.4.Attribute Definition and Data Transformation

Continuous variables were categorized primarily to improve interpretability of the resulting classification rules and facilitate bedside applicability. Age was categorized into four groups: 18–30, 31–50, 51–70, and ≥71 years10. Fever was classified into five categories: normal (<37°C), mild (37–38°C), moderate (38–39°C), high (39–40°C), and malignant (>40°C). WBC counts were categorized as normal (<10,000 cells/mm³), slightly elevated (10,000–15,000), moderately elevated (15,000–20,000), and high (>20,000). Neutrophil percentage was categorized as normal (<70%), slightly elevated (70–80%), and high (>80%), while CRP values were classified as normal (<5 mg/L) or elevated (>5 mg/L).

2.5.Imaging Findings

Ultrasound findings were classified as appendicitis, other pathology, or normal. Appendicitis on ultrasound was defined by at least one of the following: appendiceal diameter greater than 6–7 mm, wall thickening greater than 3 mm, non-compressibility, appendicolith, a blind-ended tubular structure, or peri-appendiceal fluid/abscess.

Computed tomography findings were classified in the same way, with appendicitis defined by appendiceal enlargement greater than 6 mm, mesenteric fat stranding, appendicolith, cecal thickening, adenopathy, or abscess formation.

2.6.Outcome Variables

Appendectomy status was recorded as a binary variable (yes/no). The final diagnosis (appendicitis vs. non-appendicitis) was determined according to the histopathological examination of the resected appendix. The final dataset structure is shown in Table 2.

Table 2. Data transformation and optimization

Table 2

Attribute Type Value
Age Nominal 18–30, 31–50, 51–70, and ≥71 years
Sex Nominal M, F
Fever Nominal Normal, Mild, Moderate, High, Malignant
WBC Nominal Normal, Slightly_elevated, Moderately_elevated, High
Neutrophil Nominal Normal, Slightly_elevated, High
CRP Nominal Normal, High
US Nominal Diagnose_appendicitis, Diagnose_other_pathology, Normal
CT Nominal Diagnose_appendicitis, Diagnose_other_pathology, Normal
Appendectomy Boolean Yes, No
Diagnosis Nominal Appendicitis, Not_Appendicitis

2.7.Machine Learning Model Development and Validation

Data mining analysis was performed using WEKA version 3.8 (University of Waikato, New Zealand)11. A J48 decision tree classifier, corresponding to the C4.5 algorithm, was selected because its rule-based structure allows transparent visualization of predictor interactions. Model performance was assessed using 10-fold cross-validation, with metrics averaged across all internal folds to reduce overfitting and provide robust validation.

Classification performance was comprehensively evaluated via a standard confusion matrix, calculating overall diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Furthermore, the J48 algorithm identified the most informative clinical attributes based on mathematical information gain, allowing for an objective estimation of the relative contribution of individual laboratory and imaging variables in predicting acute appendicitis.

2.8.Statistical Analysis

All predictive modeling and performance calculations were performed within the WEKA environment. Model validation and calculation of diagnostic performance metrics—including accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV)—were performed within the WEKA analytical environment based on the confusion matrix obtained from the classification results. No external validation cohort was available, and no comparison with alternative machine-learning algorithms or conventional diagnostic scores was performed in the present study.

Results

This study evaluated the classification performance of a decision tree–based data mining model for acute appendicitis using a dataset of 281 patient records with six attributes: neutrophil percentage, WBC count, CRP, ultrasound (US), computed tomography (CT), and final diagnosis. Analysis was performed in WEKA Explorer, which provided both numerical performance metrics and a visual representation of the decision tree (Figure 1). Internal validation with 10-fold cross-validation was performed to obtain a more stable estimate of model performance across dataset subsets.

The model correctly identified 150 cases of appendicitis (true positives) and 87 cases of non-appendicitis (true negatives), while 17 appendicitis cases were misclassified as non-appendicitis (false negatives) and 27 non-appendicitis cases as appendicitis (false positives).

Figure 1. Flowchart of data mining model as produced by WEKA.

This corresponded to an overall accuracy of 84.3%, with a sensitivity of 89.8%, specificity of 76.3%, positive predictive value (PPV) of 84.8%, and negative predictive value (NPV) of 83.7% (Table 3). Most misclassifications occurred in patients with atypical presentations or borderline laboratory values.

In contrast to prior reports describing male predominance, appendicitis cases in the current cohort were distributed similarly between male and female patients (male-to-female ratios ranging from 2.6:1 to 1:0.03)12. The lifetime risk of developing acute appendicitis is 8.6% in males and 6.7% in females13. Most patients presented with low-grade fever, and elevated CRP levels were common among appendicitis cases, consistent with classical clinical presentations14.

Table 3. Performance metrics of the decision tree model for predicting acute appendicitis

Table 3

Metric Value (%)
Accuracy 84.3
Sensitivity / Recall 89.8
Specificity 76.3
Positive Predictive Value (PPV) 84.8
Negative Predictive Value (NPV) 83.7

The decision tree model highlighted the diagnostic significance of imaging findings, particularly US and CT, as well as laboratory parameters such as WBC and neutrophil percentage. Integration of these clinical, laboratory, and imaging variables contributed substantially to model discrimination and reflected practical considerations of surgical decision-making.

Discussion

4.1.Epidemiology and Clinical Significance

Acute appendicitis continues to represent a major component of emergency general surgical practice across all age groups, affecting approximately 7% of individuals over their lifetime15,16. The highest incidence occurs between 10 and 30 years of age, while only 5–10% of cases are observed in elderly patients15. Although overall mortality is low (<1%), delayed or missed diagnoses can lead to severe complications, including peritonitis and sepsis, particularly in children, elderly patients, and pregnant women, where mortality may reach 50%17. Notably, approximately 15% of appendectomies are performed unnecessarily, reflecting the persistent challenges in achieving timely and accurate diagnosis18. Diagnostic errors are frequent, with 7.1–10% of adults and 4.8–15% of children initially misdiagnosed19,20, underscoring the clinical and medico-legal implications of diagnostic failure21,22.

Specific populations are particularly vulnerable to misdiagnosis. Children often present with symptoms mimicking gastroenteritis or respiratory infections (25–30%), while women of childbearing age may be misdiagnosed with gynecological or urinary conditions (33%)23. Pregnant women, although exhibiting a similar incidence, often present atypically, increasing the risk of appendicular perforation and fetal loss (up to 20%) compared with uncomplicated cases (3–5%)22. In elderly patients (>60 years), misdiagnosis is common due to comorbidities and alternative abdominal pathologies such as diverticulitis and colonic ischemia, resulting in higher complication rates (18–70%) compared to younger adults (3–29%)24,25,26. Collectively, these findings illustrate the ongoing diagnostic challenges associated with appendicitis, particularly in clinically atypical populations.

4.2.Current Diagnostic Tools and Scoring Systems

The diagnosis of acute appendicitis relies on a combination of clinical evaluation, laboratory markers, and radiological investigations. Imaging modalities, including ultrasonography (US) and computed tomography (CT), improve diagnostic precision, reduce negative appendectomy rates, and facilitate early surgical intervention27. Scoring systems such as the Alvarado, RIPASA, and AIR scores (for adults), as well as PAS, Lintula, and Tzanakis scores (for pediatric populations), assist in risk stratification but demonstrate variable sensitivity and specificity28. For example, a 2022 meta-analysis reported the Alvarado score with 72% sensitivity and 77% specificity, whereas RIPASA exhibited higher sensitivity but lower specificity. The variability observed among currently available scoring systems suggests that no single diagnostic strategy performs consistently across all clinical settings.

4.3.Data Mining in Healthcare and Rationale for This Study

In acute appendicitis, diagnosis depends on integrating clinical, laboratory, and imaging findings, yet accurate diagnostic classification at the individual-patient level remains challenging. Machine-learning approaches, including decision tree classifiers, have therefore been explored for appendicitis diagnosis and for other clinical prediction tasks29. More broadly, data mining in healthcare information systems has been used to extract clinically meaningful patterns from complex datasets and support decision-making30,31. Recent reviews also note that artificial intelligence and data science may improve healthcare delivery, although their application remains constrained by issues such as implementation complexity, privacy concerns, and variable clinical readiness32,33.

Among available analytical tools, WEKA was selected because it provides an accessible framework for exploratory classification modeling and visualization34. Comparative reviews of data mining tools for healthcare likewise identify WEKA as a flexible and accessible platform for this purpose35. In the present study, this approach was used to examine whether an interpretable decision tree could capture diagnostic patterns associated with acute appendicitis.

4.4.Study Findings and Implications

In this study, the decision tree model showed moderate internal classification performance, correctly identifying 150 cases of appendicitis and 87 non-appendicitis cases, yielding an overall accuracy of 84.3%. Laboratory parameters (WBC count, neutrophil percentage) and imaging findings (US, CT) emerged as the most informative predictors, aligning closely with established surgical decision-making. The lifetime risk of developing acute appendicitis is significant, being 8.6% in males and 6.7% in females13, and the model provides a systematic approach to support diagnostic evaluation in this population.

These findings suggest that combining clinical, laboratory, and imaging variables within a predictive model may be useful for supporting diagnostic evaluation. However, the present results should be interpreted as preliminary and exploratory. The present findings should be interpreted as preliminary evidence from an exploratory classification framework requiring further validation.

4.5.Limitations and Future Directions

This study has several limitations. First, it was conducted at a single center and included a relatively small sample, which limits generalizability. Second, the model was evaluated only with internal 10-fold cross-validation; no external validation cohort was available. Third, only one machine-learning algorithm was tested, so the present results do not show whether J48 is superior to other classifiers or to established clinical scores. Fourth, the use of ultrasound and computed tomography as predictors means that the model reflects a later stage of diagnostic workup rather than an early bedside tool. Fifth, although categorization improved interpretability, it may also have reduced the granularity of predictive information contained within continuous variables. Sixth, because the outcome was final histopathology in operated patients, the model was assessed in a selected surgical population rather than in all patients presenting with abdominal pain.

Accordingly, the findings should be interpreted as preliminary internal validation results rather than evidence of clinical effectiveness or implementation readiness.

Conclusion

In this retrospective single-center cohort, a J48 decision tree classifier demonstrated moderate internal classification performance for distinguishing appendicitis from non-appendicitis using routinely available clinical, laboratory, and imaging variables. Ultrasound findings, computed tomography findings, white blood cell count, and neutrophil percentage contributed most substantially to model classification performance. Although these findings suggest potential utility for structured diagnostic classification, external validation and comparison with established clinical scoring systems and alternative machine-learning methods remain necessary before clinical implementation.

References

  1. Channick SA. Healthcare cost containment: No longer an option but a mandate. Nev LJ. 2012;13:370.
  2. Di Saverio S, Podda M, De Simone B, et al. Diagnosis and treatment of acute appendicitis: 2020 update of the WSES Jerusalem guidelines. World J Emerg Surg. 2020;15:27. https://doi.org/10.1186/s13017-020-00306-3
  3. Kabir SA, Kabir SI, Sun R, Jafferbhoy S, Karim A. How to diagnose an acutely inflamed appendix; a systematic review of the latest evidence. Int J Surg. 2017;40:155-162. https://doi.org/10.1016/j.ijsu.2017.03.013
  4. Sandell E, Berg M, Sandblom G, et al. Surgical decision-making in acute appendicitis. BMC Surg. 2015;15:69. https://doi.org/10.1186/s12893-015-0053-x
  5. Snyder MJ, Guthrie M, Cagle S. Acute Appendicitis: Efficient Diagnosis and Management. Am Fam Physician. 2018;98(1):25-33.
  6. Sartelli M, Baiocchi GL, Catena GL, et al. Prospective Observational Study on Acute Appendicitis Worldwide (POSAW). World J Emerg Surg. 2018;13:19. https://doi.org/10.1186/s13017-018-0179-0
  7. Sharma P, Jain A, Shankar G, Jinkala S, Kumbhar US, Shamanna SG. Diagnostic accuracy of Alvarado, RIPASA and Tzanakis scoring system in acute appendicitis: A prospective observational study. Trop Doct. 2021;51(4):475-481. https://doi.org/10.1177/00494755211030165
  8. Saeed S, Shaikh A, Memon M, Naqvi SM. Impact of Data Mining Techniques to Analyze Health Care Data. J Med Imaging Health Inform. 2018;8:682-690. https://doi.org/10.1166/jmihi.2018.2385
  9. Walczak S, Scharf J. Reducing surgical patient cost through use of an artificial neural network to predict transfusion requirements. Decis Support Syst. 2000;30:125-138. https://doi.org/10.1016/S0167-9236(00)00093-2
  10. Mohanty S, Satapathy MP, Sahu DN. Study of incidence and demographic profile of inguinal hernia in adults. Int J Res Med Sci. 2018;6(9):2969-2973.
  11. Jagtap SB, Kodge BG. Census Data Mining and Data Analysis using WEKA. ICETSTM – 2013 International Conference in Emerging Trends in Science, Technology and Management. Singapore; 2013.
  12. Kollias TF, Gallagher CP, Albaashiki A, Burle VS, Slouha E. Sex Differences in Appendicitis: A Systematic Review. Cureus. 2024;16(5):e60055. https://doi.org/10.7759/cureus.60055
  13. Hatem F, Baig H, Khaldas F, Lucocq J. Negative Appendicectomy Rates in Females of Childbearing Age: A Retrospective Analysis and Literature Review. Cureus. 2022;14(7):e27412. https://doi.org/10.7759/cureus.27412
  14. Moris D, Paulson EK, Pappas TN. Diagnosis and Management of Acute Appendicitis in Adults: A Review. JAMA. 2021;326(22):2299-2311. https://doi.org/10.1001/jama.2021.20502
  15. Omari AH, Khammash MR, Qasaimeh GR, Shammari AK, Yaseen MK, Hammori SK. Acute appendicitis in the elderly: risk factors for perforation. World J Emerg Surg. 2014;9(1):6. https://doi.org/10.1186/1749-7922-9-6
  16. Krzyzak M, Mulrooney SM. Acute Appendicitis Review: Background, Epidemiology, Diagnosis, and Treatment. Cureus. 2020;12:e8562. https://doi.org/10.7759/cureus.8562
  17. Bhargavi VA, Abhijeeth DS, Amar DN. The Accuracy of Leucocyte Count in Diagnosing Acute Appendicitis. Int J Med Pharm Res. 2003;4(3):14-19.
  18. Flum DR, Morris A, Koepsell T, Dellinger EP. Has Misdiagnosis of Appendicitis Decreased Over Time? A Population-Based Analysis. JAMA. 2001;286(14):1748-1753. https://doi.org/10.1001/jama.286.14.1748
  19. Weinberger H, Zeina AR, Ashkenazi I. Misdiagnosis of Acute Appendicitis in the Emergency Department: Prevalence, Associated Factors, and Outcomes According to the Patients’ Disposition. Ochsner J. 2023. https://doi.org/10.31486/toj.23.0051
  20. Naiditch JA, Lautz TB, Daley S, Pierce MC, Reynolds M. The implications of missed opportunities to diagnose appendicitis in children. Acad Emerg Med. 2013;20(6):592-596. https://doi.org/10.1111/acem.12144
  21. Chang YJ, Chao HC, Kong MS, Hsia SH, Yan DC. Misdiagnosed Acute Appendicitis in Children in the Emergency Department. Chang Gung Med J. 2010;33(5):551-556.
  22. Raveenthiran V, Darun N, Shobini S. Neglected Appendicitis. In: Weledji E, editor. Appendicitis Causes and Treatment. IntechOpen; 2023. https://doi.org/10.5772/intechopen.1001119
  23. Rothrock SG, Green SM, Dobson M, Colucciello SA, Simmons CM. Misdiagnosis of appendicitis in nonpregnant women of childbearing age. J Emerg Med. 1995;13(1):1-8. https://doi.org/10.1016/0736-4679(94)00104-9
  24. Baek HN, Jung YH, Hwang YH. Laparoscopic versus open appendectomy for appendicitis in elderly patients. J Korean Soc Coloproctol. 2011;27(5):241-245. https://doi.org/10.3393/jksc.2011.27.5.241
  25. Min LQ, Lu J, He HY. Clinical significance of appendicoliths in elderly patients over eighty years old undergoing emergency appendectomy: A single-center retrospective study. World J Gastrointest Surg. 2024;16(11):3453-3462. https://doi.org/10.4240/wjgs.v16.i11.3453
  26. Gaitur A. Clinical effectiveness study of the new diagnostic score of acute appendicitis in the elderly. Mold J Health Sci. 2023;10(3):25-34. https://doi.org/10.52645/MJHS.2023.3.04
  27. Incesu L, Coskun A, Selcuk MB, Akan H, Sozubir S, Bernay F. Acute Appendicitis: MR imaging and sonographic correlation. AJR Am J Roentgenol. 1997;168(3):669-674. https://doi.org/10.2214/ajr.168.3.9057519
  28. Favara G, Maugeri A, Barchitta M, Ventura A, Basile G, Agodi A. Comparison of RIPASA and ALVARADO scores for risk assessment of acute appendicitis: A systematic review and meta-analysis. PLoS ONE. 2022;17(9):e0275427. https://doi.org/10.1371/journal.pone.0275427
  29. Akmeşe Ö, Dogan G, Kör H, Erbay H, Demir E. The Use of Machine Learning Approaches for the Diagnosis of Acute Appendicitis. Emerg Med Int. 2020;2020:7306435. https://doi.org/10.1155/2020/7306435
  30. Shahin A, Moudani W, Chakik F, Khalil M. Data Mining In Healthcare Information Systems: Case Studies In Northern Lebanon. 2014 3rd International Conference on e-Technologies and Networks for Development (ICeND). 2014. https://doi.org/10.1109/ICeND.2014.6991370
  31. Gunturi Subrahmanya SV, Shetty DK, Patil V, et al. The role of data science in healthcare advancements: Applications, benefits, and future prospects. Ir J Med Sci. 2022;191(4):1473-1483. https://doi.org/10.1007/s11845-021-02730-z
  32. David BO, Aanuoluwapo CDO, Ojima ZW, Akinsola JA, Temitope A, Jonathan L. Artificial intelligence in healthcare delivery: Prospects and pitfalls. J Med Surg Public Health. 2024;3:100108. https://doi.org/10.1016/j.glmedi.2024.100108
  33. Williamson SM, Prybutok V. Balancing Privacy and Progress: A Review of Privacy Challenges, Systemic Oversight, and Patient Perceptions in AI-Driven Healthcare. Appl Sci. 2023;14(2):675. https://doi.org/10.3390/app14020675
  34. Witten I, Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P. The WEKA data mining software: An update. SIGKDD Explor. 2009;11:10-18. https://doi.org/10.1145/1656274.1656278
  35. Santos-Pereira J, Gruenwald L, Bernardino J. Top data mining tools for the healthcare industry. J King Saud Univ Comput Inf Sci. 2022;34:4968-4982. https://doi.org/10.1016/j.jksuci.2021.06.002

Cite this article

Ali Bekraki, Ahmad Shahin. Interpretable Decision Tree Model for Classification of Acute Appendicitis Using Clinical, Laboratory, and Imaging Data: A Retrospective Single-Center Study.. Journal of Cukurova Anesthesia and Surgical Sciences. 9(2):506-514. https://doi.org/10.36516/jocass.1958579

Scroll to Top