medigraphic.com
ISSN 2992-7749 (Electronic)
ISSN 2992-7897 (Print)
Órgano Oficial de difusión científica de la Asociación Mexicana de Cirujanos de Columna A. C. y de la Sociedad Iberolatinoamericana de Columna. SILACO.

Indizada en:

Scielo

2026, Number 4

<< Back

Cir Columna 2026; 4 (4)

Systematic review assisted by artificial intelligence: usefulness, challenges and recommendations

Salcido RMV

Language: Spanish
References: 17
Page: 349-356
PDF size: 739.43 Kb.


ABSTRACT

Introduction: the systematic review (SR) represents the highest level of scientific evidence. Its elaboration, however, is a complex process with high demands on time and resources. Artificial intelligence (AI) emerges as a tool with the potential to transform each stage of this process. Objective: to critically evaluate the role of artificial intelligence as an assistive tool in the elaboration of systematic reviews, identifying its methodological and ethical advantages and limitations, and proposing criteria for its responsible implementation based on the PRISMA-P 2015 checklist. Material and methods: a narrative review of the literature was conducted in PubMed, Cochrane, Scopus and Web of Science on the use of AI in systematic reviews and meta-analyses, complemented by analysis of documents from experts in research methodology, including Jiménez Ávila and collaborators. Results: AI offers significant advantages in processing large volumes of information, reducing operational bias, and automating search, screening and data extraction. Relevant challenges were identified: algorithmic bias, hallucinations, lack of transparency, limited reproducibility, and the absence of consolidated methodological standards. Tools such as ASReview, Covidence, Elicit, GRADEpro and Rayyan demonstrated proven utility at different stages of the process. Conclusions: AI improves efficiency and can contribute to the quality of systematic reviews when its use is supervised, declared and methodologically justified. It does not replace the scientific judgment of the researcher. Its incorporation should be progressive, critical and subject to explicit ethical and transparency standards.


REFERENCES

  1. Higgins JPT, Thomas J, Chandler J, et al. CochraneHandbook for Systematic Reviews of Interventions,Version 6.3. Cochrane; 2022. Disponible en: https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current

  2. Cumpston M, Li T, Page MJ, Chandler J, WelchVA, Higgins JP, et al. Updated guidance for trustedsystematic reviews: a new edition of the CochraneHandbook for Systematic Reviews of Interventions.Cochrane Database Syst Rev. 2019; 10: ED000142.doi: 10.1002/14651858.ED000142.

  3. Bornmann L, Mutz R. Growth rates of modernscience: a bibliometric analysis based on the numberof publications and cited references. J Assoc InfSci Technol. 2015; 66: 2215-2222. doi: 10.1002/asi.23329.

  4. Elsevier. Scopus Content Coverage Guide. Amsterdam:Elsevier B.V.; 2023. Disponible en: www.elsevier.com/solutions/scopus/how-scopus-works/content

  5. Liu PR, Lu L, Zhang JY, Huo TT, Liu SX, Ye ZW.Application of artificial intelligence in medicine: anoverview. Curr Med Sci. 2021; 41(6): 1105-1115. doi:10.1007/s11596-021-2474-3.

  6. Jiménez ÁJM. La revolución de la inteligencia artificialgenerativa en el área de la salud y la cirugía de columna.Cir Columna. 2025; 3: 76-77. doi: 10.35366/119615.

  7. Van Dijk SHB, Brusse-Keizer MGJ, Bucsán CC, Van DerPalen J, Doggen CJM, Lenferink A. Artificial intelligencein systematic reviews: promising when appropriatelyused. BMJ Open. 2023; 13: e072254. doi: 10.1136/bmjopen-2023-072254.

  8. Von Groote T, Ghoreishi N, Bjorklund M, PorschenC, Puljak L. Exponential growth of systematic reviewsassessing artificial intelligence studies in medicine: challenges and opportunities. Syst Rev. 2022; 11: 132.doi: 10.1186/s13643-022-01984-7.

  9. Grzybowski A, Jin K, Wu H. Challenges of artificialintelligence in medicine and dermatology. ClinDermatol. 2024; 42: 210-215. doi: 10.1016/j.clindermatol.2023.12.013.

  10. Alkaissi H, McFarlane SI. Artificial hallucinations inChatGPT: implications in scientific writing. Cureus. 2023;15: e35179. doi: 10.7759/cureus.35179.

  11. Bernard N, Sagawa Y Jr, Bier N, Lihoreau T, PazartL, Tannou T. Using artificial intelligence for systematicreview: the example of Elicit. BMC Med Res Methodol.2025; 25: 75. doi: 10.1186/s12874-025-02528-y.

  12. Crossnohere NL, Elsaid M, Paskett J, Bose-Brill S,Bridges JFP. Guidelines for artificial intelligence inmedicine: literature review and content analysis offrameworks. J Med Internet Res. 2022; 24: e36823. doi:10.2196/36823.

  13. Henzler D, Schmidt S, Kocar A, et al. Healthcareprofessionals’ perspectives on artificial intelligence inpatient care: a systematic review. BMC Health ServRes. 2025; 25: 416. doi: 10.1186/s12913-025-12664-2

  14. Shamseer L, Moher D, Clarke M, Ghersi D, LiberatiA, Petticrew M, et al. Preferred reporting items forsystematic review and meta-analysis protocols(PRISMA-P) 2015: elaboration and explanation. BMJ.2015; 349: g7647. doi: 10.1136/bmj.g7647.

  15. Thomas J, McDonald S, Noel-Storr A, Shemilt I,Elliott J, Mavergames C, et al. Machine learningreduced workload with minimal risk of missing studies:development and evaluation of a randomized controlledtrial classifier for Cochrane Reviews. J Clin Epidemiol.2021; 133: 140-151. doi: 10.1016/j.jclinepi.2020.11.003.PMID: 33171275.

  16. Marshall IJ, Kuiper J, Banner E, Wallace BC. Automatingbiomedical evidence synthesis: RobotReviewer. J ProcConf Assoc Comput Linguist Meet. 2017; 2017: 7-12.doi: 10.18653/v1/P17-4002.

  17. Jiménez ÁJM, Negrete IJ, Hyun JS. La inteligenciaartificial en la investigación en el ámbito de la salud:desafíos y oportunidades. Cir Columna. 2025; 3: 139-145. doi: 10.35366/119625.




2020     |     www.medigraphic.com