Basınç Yaralanmasının Tespit ve Sınıflandırılmasında Derin Öğrenme Modelinin Etkinliğinin Değerlendirilmesi

Öz

Amaç: Araştırma basınç yaralanmalarının tespit ve sınıflandırılmasında derin öğrenme modelinin hemşirelerin bilgi ve memnuniyet düzeylerine etkisini belirlemek amacıyla yürütüldü. Yöntem: Randomize kontrollü tasarımla yürütülen bu çalışmanın evrenini bir vakıf üniversitesi hastanesinde Mart–Nisan 2022 tarihlerinde yoğun bakım, dahiliye ve cerrahi kliniklerde çalışan ve çalışmaya gönüllü hemşireler oluşturdu. Örneklemi ise örneklem kriterlerine uyan toplam 60 (30 deney ve 30 kontrol) hemşire oluşturdu. Araştırma verileri, Yapılandırılmış Hemşire Tanıtım Formu, Modifiye Pieper Basınç Yarası Bilgi Testi ve Hemşire Memnuniyet Skalası kullanılarak toplandı. Araştırma verileri SPSS 25.0 programında analiz edildi. Bulgular: Deney grubu hemşirelerin yaş ortalaması 25,67±7,27, kontrol grubu 25,10±3,47 olarak tespit edildi. Deney ve kontrol grubu hemşirelerin %50‘si sağlık meslek lisesi mezunu, %40’ı cerrahi servislerde çalışmaktadır. Hemşirelerin eğitim sonrası bilgi sınavı (sontest) puanları karşılaştırıldığında; deney grubunun ortalama puanı 39,36±1,88, kontrol grubunun 33,30±1,68 olarak belirlendi. Deney grubunun eğitim sonrası bilgi düzeyi kontrol grubundan istatistiksel olarak anlamlı derecede yüksek bulundu (P<,05). Basınç yaralanması risk değerlendirme ve evre tespit etme başarısı incelendiğinde deney grubunun derin öğrenme modeliyle %97 başarıyla risk değerlendirebildiği ve %89 tahmin doğrulamayla yara evresi belirleyebildiği tespit edildi. Kontrol grubunun Braden bası yarası risk değerlendirme ölçeği ile hastaların risk düzeylerini 13,83±4,67 ile orta düzeyde belirlediği saptandı. Deney grubunun basınç yarası risk değerlendirme ve evre tahmin etme düzeyleri kontrol grubundan istatistiksel olarak anlamlı derecede yüksek bulundu (P<,05). Araştırmaya katılan hemşirelerin uygulanan eğitimden memnuniyet düzeyleri incelendiğinde; deney grubunun puan ortalaması 24,60±0,96 ve kontrol grubunun 20,93±0,63 olarak belirlendi. Deney grubunun eğitimden memnuniyet düzeyi kontrol grubundan istatistiksel olarak anlamlı derecede yüksek bulundu (P<,05). Sonuç: Yapay zeka teknolojisiyle basınç yaralanması tespit ve sınıflandırmasının geleneksel yönteme göre daha başarılı olduğu tespit edildi.

Anahtar Kelimeler

• Basınç yaralanması
• derin öğrenme
• yapay zeka

Evaluation of the Effectiveness of Deep Learning Model in Detection and Classification of Pressure Injury

Abstract

Objective: The study was conducted to determine the effect of the deep learning model on the knowledge and satisfaction levels of nurses in the detection and classification of pressure injuries. Method: The population of this randomized controlled trial consisted of nurses working in intensive care, internal medicine, and surgical clinics at a foundation university hospital between March and April 2022 who voluntarily participated in the study. The sample consisted of a total of 60 (30 experimental and 30 control) nurses who met the sample criteria. The research data were collected using the Structured Nurse Introduction Form, Modified Pieper Pressure Injury Knowledge Test and Nurse Satisfaction Scale.The research data were analyzed in the SPSS 25.0 program. Results: The mean age of the nurses in the experimental group was determined as 25.67±7.27, and the control group as 25.10±3.47. 50% of the nurses in the experimental and control groups graduated from health vocational high schools, and 40% of them worked in surgical services. When the nurses' post-training knowledge exam (post-test) scores were compared; the mean score of the experimental group was determined as 39.36±1.88 and the control group as 33.30±1.68. The post-training knowledge level of the experimental group was found to be statistically significantly higher than the control group (P<.05). When the success of the pressure injury risk assessment and stage determination was examined, it was determined that the experimental group was able to assess the risk with 97% success with the deep learning model and determine the wound stage with 89% prediction verification. It was determined that the control group determined the patients' risk levels with the Braden pressure injury risk assessment scale at a moderate level with 13.83±4.67 and were 50% successful in stage estimation. The evaluation and stage estimation levels were found to be statistically significantly higher than the control group (P<.05). When the satisfaction levels of the nurses participating in the study with the applied training were examined; the average score of the experimental group was determined as 24.60±0.96 and the control group as 20.93±0.63. The satisfaction level of the experimental group with the training was found to be statistically significantly higher than the control group (P<.05). Conclusion: It was determined that pressure injury detection and classification with artificial intelligence technology was more successful than the traditional method.

Keywords

• Artificial intelligence
• deep learning
• pressure injury.

References

1. 1. Alderden J, Pepper GA, Wilson A, et al. Predicting pressure injury in critical care patients: A machine-learning model. Am J Crit Care. 2018;27(6):461–468.
2. 2. Edsberg LE, Black JM, Goldberg M, et al. Revised National Pressure Ulcer Advisory Panel pressure injury staging system. J Wound Ostomy Continence Nurs. 2016;43(6):585–585.
3. 3. Ferris A, Price A, Harding K. Pressure ulcers in patients receiving palliative care: A systematic review. Palliat Med. 2019;33(7):770–782.
4. 4. Kottner J, Cuddigan J, Carville K, et al. Pressure ulcer/injury classification today: An international perspective. J Tissue Viability. 2020;12(1):1–10.
5. 5. Martinengo L, Yeo NJY, Tang ZQ, et al. Digital education for the management of chronic wounds in health care professionals: Protocol for a systematic review by the Digital Health Education Collaboration. JMIR Res Protoc. 2019;8(3):12–48.
6. 6. McGinnis E, Brown S, Collier H, et al. Pressure relieving support surfaces: A randomised evaluation 2 (PRESSURE 2) photographic validation sub-study: Study protocol for a randomised controlled trial. Trials. 2017;18(1):1–10.
7. 7. Moore ZE, Patton D. Risk assessment tools for the prevention of pressure ulcers. Cochrane Database Syst Rev. 2019;1(1):1–10.
8. 8. Ören N. Investigation of Nurses' Knowledge and Stages of Pressure Ulcer Diagnosis. Master’s Thesis. Zonguldak Bülent Ecevit University, Institute of Health Sciences; 2019.

9. 9. Raju D, Su X, Patrician PA, Loan LA, McCarthy MS. Exploring factors associated with pressure ulcers: A data mining approach. Int J Nurs Stud. 2015;52(1):102–111.
10. 10. Van Den Oord A, Dieleman S, Schrauwen B. Deep content-based music recommendation. Adv Neural Inf Process Syst. 2013;26(1):1–15.
11. 11. Yap TL, Kennerly SM, Ly K. Pressure injury prevention: Outcomes and challenges to use of resident monitoring technology in a nursing home. J Wound Ostomy Continence Nurs. 2019;46(3):207–213.
12. 12. Gökalp MG, Üzer MA. Nursing care in the age of artificial intelligence. J Health Sci Univ Nurs. 2024;6(1):89–94.
13. 13. Yilmaz A. Artificial intelligence. 6th ed. Istanbul: Kodlab Publishing House; 2017.
14. 14. Ahmed SK. Artificial intelligence in nursing: Current trends, possibilities and pitfalls. J Med Surg Public Health. 2024;3(1):72–100.
15. 15. Rony MKK, Parvin MR, Ferdousi S. Advancing nursing practice with artificial intelligence: Enhancing preparedness for the future. Nurs Open. 2024;11(1):25–35.
16. 16. De Gagne JC, Hwang H, Jung D. Cyberethics in nursing education: Ethical implications of artificial intelligence. Nurs Ethics. 2024;31(6):1021–1030.
17. 17. Pınar R, Oğuz S. Testing the reliability and validity of the Norton and Braden Pressure Ulcer Assessment Scales in the same bedridden patient group. In: 6th National Nursing Congress with International Participation. Istanbul: Damla Publishing House; 1998. p.172–175.
18. 18. Topaz M, Peltonen LM, Michalowski M, et al. The ChatGPT effect: Nursing education and generative artificial intelligence. J Nurs Educ. 2024;1(1):1–4.
19. 19. Boussina A, Shashikumar SP, Malhotra A, et al. Impact of a deep learning sepsis prediction model on quality of care and survival. NPJ Digit Med. 2024;7(1):1–14.
20. 20. Wu XY. Exploring the effects of digital technology on deep learning: A meta-analysis. Educ Inf Technol. 2024;29(1):425–458.
21. 21. Rasul MG, et al. Future prediction for nurse care activities using deep learning-based multi-label classification. In: Human Activity and Behavior Analysis. Boca Raton (FL): CRC Press; 2024. p.377–387.
22. 22. Khedr AM. Enhancing supply chain management with deep learning and machine learning techniques: A review. J Open Innov Technol Mark Complex. 2024;1(1):37–39.
23. 23. Yaghoubi E, Khamees A, Razmi D, Lu T. A systematic review and meta-analysis of machine learning, deep learning, and ensemble learning approaches in predicting EV charging behavior. Eng Appl Artif Intell. 2024;135(10):87–89.
24. 24. Swerdlow M, Guler O, Yaakov R, Armstrong DG. Simultaneous segmentation and classification of pressure injury image data using Mask-R-CNN. Comput Math Methods Med. 2023;2023(1):3858997.
25. 25. Demircan F, Toz M, Yücedağ I. Analysis of pressure ulcer formation risk in different regions of the human body. Konuralp J Math. 2016;6(2):233–239.
26. 26. Campitiello F, Mancone M, Corte AD, Guerniero R, Canonico S. Expanded negative pressure wound therapy in healing diabetic foot ulcers: A prospective randomised study. J Wound Care. 2021;30(2):121–129.
27. 27. Gul A, Andsoy II, Ozkaya B, Zeydan A. A descriptive, cross-sectional survey of Turkish nurses’ knowledge of pressure ulcer risk, prevention, and staging. Ostomy Wound Manage. 2017;63(6):40–46.
28. 28. Yilmaz A, Kızıl H, Kaya U, Çakır R, Demiral M. Prediction and classification of pressure injuries by deep learning. Health Problems of Civilization. 2021;15(4):328–335.
29. 29. Kim J, Lee C, Choi S, et al. Augmented decision-making in wound care: Evaluating the clinical utility of a deep-learning model for pressure injury staging. Int J Med Inform. 2023;180(1):52–66.
30. 30. Seo S, Kang J, Eom IH, et al. Visual classification of pressure injury stages for nurses: A deep learning model applying modern convolutional neural networks. J Adv Nurs. 2023;79(1):3047–3056.
31. 31. Jiang M, Ma Y, Guo S, et al. Using machine learning technologies in pressure injury management: Systematic review. JMIR Med Inform. 2021;9(3):251–273.
32. 32. Shepherd MM, Wipke-Tevis DD, Alexander GL. Analysis of qualitative interviews about the impact of information technology on pressure ulcer prevention programs. J Wound Ostomy Continence Nurs. 2015;42(3):235–241.
33. 33. Garcia-Zapirain B, Sierra-Sosa D, Ortiz D, Isaza-Monsalve M, Elmaghraby A. Efficient use of mobile devices for quantification of pressure injury images. Health Technol. 2018;26(1):269–280.
34. 34. Ng ZQP, Ling LYJ, Chew HSJ, Lau Y. The role of artificial intelligence in enhancing clinical nursing care: A scoping review. J Nurs Manag. 2022;30(8):3654–3674.
35. 35. Reifsnider E. Nursing research, practice, education, and artificial intelligence: What is our future? Res Nurs Health. 2023;46(6):564–565.
36. 36. Jayakumar P, Moore MG, Furlough KA, et al. Comparison of an AI-enabled decision aid vs educational material on outcomes in adults with knee osteoarthritis. JAMA Netw Open. 2021;4(2):e2037107.
37. 37. Pei J, Guo X, Tao H, et al. Machine learning-based prediction models for pressure injury: A systematic review and meta-analysis. Int Wound J. 2023;20(10):4328–4339.
38. 38. Liu H, Hu J, Zhou J, Yu R. Application of deep learning to pressure injury staging. J Wound Care. 2024;33(5):368–378.
39. 39. Wu SC, Li YJ, Chen HL, et al. Using artificial intelligence for the early detection of micro-progression of pressure injuries. Stud Health Technol Inform. 2022;290(1):1016–1017.
40. 40. Alderden J, Johnny J, Brooks KR, et al. Explainable artificial intelligence for early prediction of pressure injury risk. Am J Crit Care. 2024;33(5):373–381.
41. 41. Anderson C, Bekele Z, Qiu Y, Tschannen D, Dinov ID. Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence. BMC Med Inform Decis Mak. 2021;21(1):253–255.
42. 42. Song W, Kang MJ, Zhang L, et al. Predicting pressure injury using nursing assessment phenotypes and machine learning methods. J Am Med Inform Assoc. 2021;28(4):759–765.