Research Article
BibTex RIS Cite

Makine Öğrenmesi Algoritmaları Kullanılarak Vücut Analizi ile Uyku Apnesi Teşhisi

Year 2021, Volume: 2 Issue: 1, 6 - 10, 01.06.2021

Abstract

Amaç
Bu çalışmada vücut analizi (TANİTA) verileri ile apne-hipopne indeksine (AHİ) göre obstrüktif uyku apne sendromu (OSAS) şiddetinin tahmin edilmesi amaçlandı.
Metod
Bu çalışmada OSAS tanısı konulan ve diyet polikliniğine başvuran 239 adet hastanın, her biri için 23 adet vücut analizi verisi kullanılmıştır. Farklı vücut analizi verisi seçimi algoritmaları çalıştırılarak, veri madenciliği modellemeleri yapılmıştır. Sonuç olarak en başarılı sonuçların elde edildiği metod; Korelasyon Tabanlı Öznitelik Seçimi yöntemidir (CFS). CFS yöntemine göre bu çalışmada dört adet öznitelik, sınıflandırma işleminde en başarılı sonucu verdirmiştir. Seçilen öznitelikler Kilo, Metabolizma Yaşı, Hücre Dışı Sıvı ve Bel çevresi olmuştur. Bu çalışmada altı adet sınıflandırma yöntemi (Bayes Sınıflandırıcı Algoritması, Destek Vektör Makineleri, K* Algoritması, Reptree Algoritması, ZeroR Algoritması, Yapay Sinir Ağları) kullanılarak sonuçların başarısı karşılaştırılmıştır.
Bulgular
Her bir sınıflandırma yönteminden elde edilen en iyi sonuçlar verilmiştir. Burada tüm gruplar için AHİ’yi tahmin etme oranı %65 olarak bulundu ve en iyi tahmin yapay zeka modelinde oldu. Modeller kullanılarak AHI’ye göre yapılan sınıflandırmada; hafif OSAS 153 hasta, orta OSAS 43 hasta, ağır OSAS 43 hasta bulundu. DVM %100, ZeroR %100, YSA %100 90,85 olarak bulunmuştur. Burada en iyi hafif OSAS’ı tahmin etmede etkili olmuştur.

Sonuç
Sonuçlar incelendiğinde sınıflandırma yöntemleri kullanılarak vücut analizi verileri ile AHI tahmininin yapılabileceği öngörülmektedir. En iyi sınıflandırma yöntemi Yapay Sinir Ağları, ZeroR ve DVM modelleri ile elde edilmiştir. Sonuçların daha iyi tahmin yüzdesine sahip olabilmesi için daha geniş hasta sayıları ile yeni çalışmalara ihtiyaç olduğunu düşünmekteyiz.

References

  • 1. O’Donnell CP, Schwartz AP, Smith PL. Upper airway collapsibility. Am J Respir Crit Care Med 2000; 162:1606–1607.
  • 2. Wolk R, Shamsuzzaman AS, Somers VK. Obesity, sleep apnoea and hypertension. Hypertension 2003; 42(6):1067–1074.
  • 3. Cistulli PA. Craniofacial abnormalities in obstructive sleep apnoea: implications for treatment. Respirology 1996; 3:167–174.
  • 4. Kyzer S, Charuzi I. Obstructive sleep apnoea in obese. World J Surg 1998; 22:998–1001.
  • 5. Kansanen M, VYSAinen E, Tuunainen A et al. The eVect of a very low-calorie diet-induced weight loss on the severity of obstructive sleep apnoea and autonomic nervous function in obese patients with obstructive sleep apnoea syndrome. Clin Physiol 1998; 18:377–385. 6. Sampol G, Munoz X, Sagales MT et al. Long-term eYcacy of dietary weight loss in sleep apnoea/hypopnoea syndrome. Eur Respir J 1998; 18:377–385.
  • 7. Rollheim J, Osnes T, Miljeteig H. The relationship between obstructive sleep apnoea and body mass index. Clin Otolaryngol 1997; 22:419–422.
  • 8. Yu X, Fujimoto K, Urushibata K et al. Cephalometric analysis in obese and nonobese patients with obstructive sleep apnea syndrome. Chest 2003; 124:212–218.
  • 9. Rollheim J, Osnes T, Miljeteig H. The relationship between obstructive sleep apnoea and body mass index. Clin Otolaryngol 1997; 22:419–422.
  • 10. Busetto L, Enzi G, Inelmen EM et al. Obstructive sleep apnea syndrome in morbid obesity: eVects of intragastric baloon. Chest 2005; 128:618–623.
  • 11. Whittle A, Marshall I, Mortimore I et al. Neck soft tissue and fat distribution: comparison between normal men and women by magnetic resonance imaging. Thorax 1999; 54:323–328.
  • 12. Wolk R, Shamsuzzaman AS, Somers VK. Obesity, sleep apnoea and hypertension. Hypertension 2003; 42(6):1067–1074.
  • 13. Schafer H, Pauleit D, Sudhop T et al. Body fat distribution, serum leptin, and cardiovascular risk factors in men with obstructive sleep apnoea. Chest 2002; 122:829–839.
  • 14. Rollheim J, Osnes T, Miljeteig H. The relationship between obstructive sleep apnoea and body mass index. Clin Otolaryngol 1997; 22:419–422.
  • 15. Akita Y, Kawakatsu K, Hattori H et al. Posture of patients with sleep apnoea during sleep. Acta otolaryngol 2003; 550:41–45.
  • 16. Busetto L, Enzi G, Inelmen EM et al. Obstructive sleep apnea syndrome in morbid obesity: eVects of intragastric baloon. Chest 2005; 128:618–623.
  • 17. Lofaso F, Coste A, d’Ortho MP et al. Nasal obstruction as a risk factor for sleep apnoea syndrome. Eur Respir J 2000; 16(4):639–643.
  • 18. Yu X, Fujimoto K, Urushibata K et al. Cephalometric analysis in obese and nonobese patients with obstructive sleep apnea syndrome. Chest 2003; 124:212–218.

The diagnosis of OSAS with Body Analysis using Machine Learning Algorithm

Year 2021, Volume: 2 Issue: 1, 6 - 10, 01.06.2021

Abstract

Aim
In this study, it was aimed to predict the severity of obstructive sleep apnea syndrome (OSAS) according to apnea-hypopnea index (AHI) by body analysis (TANITA) data.
Material and Method
Twenty-three parameters of body analysis were used for each patient who had been diagnosed as OSAS and admitted to diet polyclinic. Data mining modeling has been done by running different body analysis data selected algorithms. As a result, the most successful result was obtained by Correalation – based Feature Selection (CFS). According to the CFS method, four features resulted successfully in the classification process in this study. Selected features included of model were Weight, Metabolism Age, Extracellular Fluid and Waist circumference. The success of the results was compared by using six classification methods (Naive Bayes Classifier Algorithm, Support Vector Machine, K* Algorithm, Reptree Algorithm, ZeroR Algorithm, Artificial Neural Network) in this study.
Results
The best results obtained from each classification method. The estimated ratio of AHI for all groups was 65% for YSA model. The classification was done according to AHI by using the models; mild OSAS (153 patient), moderate OSAS (43 patient), severe OSAS (43 patient), and powerfull estimate are shown DVM %100, ZeroR %100, YSA %100 90,85 for mild OSAS.
Conclusion
When the results are examined, it is predicted that the AHI estimation can be performed by using body composition data with the classification methods. The best classification method was obtained with Artificial İntelligence Networks, ZeroR and DVM models. We think that new studies are needed with larger populations to obtain better estimate percentages.

References

  • 1. O’Donnell CP, Schwartz AP, Smith PL. Upper airway collapsibility. Am J Respir Crit Care Med 2000; 162:1606–1607.
  • 2. Wolk R, Shamsuzzaman AS, Somers VK. Obesity, sleep apnoea and hypertension. Hypertension 2003; 42(6):1067–1074.
  • 3. Cistulli PA. Craniofacial abnormalities in obstructive sleep apnoea: implications for treatment. Respirology 1996; 3:167–174.
  • 4. Kyzer S, Charuzi I. Obstructive sleep apnoea in obese. World J Surg 1998; 22:998–1001.
  • 5. Kansanen M, VYSAinen E, Tuunainen A et al. The eVect of a very low-calorie diet-induced weight loss on the severity of obstructive sleep apnoea and autonomic nervous function in obese patients with obstructive sleep apnoea syndrome. Clin Physiol 1998; 18:377–385. 6. Sampol G, Munoz X, Sagales MT et al. Long-term eYcacy of dietary weight loss in sleep apnoea/hypopnoea syndrome. Eur Respir J 1998; 18:377–385.
  • 7. Rollheim J, Osnes T, Miljeteig H. The relationship between obstructive sleep apnoea and body mass index. Clin Otolaryngol 1997; 22:419–422.
  • 8. Yu X, Fujimoto K, Urushibata K et al. Cephalometric analysis in obese and nonobese patients with obstructive sleep apnea syndrome. Chest 2003; 124:212–218.
  • 9. Rollheim J, Osnes T, Miljeteig H. The relationship between obstructive sleep apnoea and body mass index. Clin Otolaryngol 1997; 22:419–422.
  • 10. Busetto L, Enzi G, Inelmen EM et al. Obstructive sleep apnea syndrome in morbid obesity: eVects of intragastric baloon. Chest 2005; 128:618–623.
  • 11. Whittle A, Marshall I, Mortimore I et al. Neck soft tissue and fat distribution: comparison between normal men and women by magnetic resonance imaging. Thorax 1999; 54:323–328.
  • 12. Wolk R, Shamsuzzaman AS, Somers VK. Obesity, sleep apnoea and hypertension. Hypertension 2003; 42(6):1067–1074.
  • 13. Schafer H, Pauleit D, Sudhop T et al. Body fat distribution, serum leptin, and cardiovascular risk factors in men with obstructive sleep apnoea. Chest 2002; 122:829–839.
  • 14. Rollheim J, Osnes T, Miljeteig H. The relationship between obstructive sleep apnoea and body mass index. Clin Otolaryngol 1997; 22:419–422.
  • 15. Akita Y, Kawakatsu K, Hattori H et al. Posture of patients with sleep apnoea during sleep. Acta otolaryngol 2003; 550:41–45.
  • 16. Busetto L, Enzi G, Inelmen EM et al. Obstructive sleep apnea syndrome in morbid obesity: eVects of intragastric baloon. Chest 2005; 128:618–623.
  • 17. Lofaso F, Coste A, d’Ortho MP et al. Nasal obstruction as a risk factor for sleep apnoea syndrome. Eur Respir J 2000; 16(4):639–643.
  • 18. Yu X, Fujimoto K, Urushibata K et al. Cephalometric analysis in obese and nonobese patients with obstructive sleep apnea syndrome. Chest 2003; 124:212–218.
There are 17 citations in total.

Details

Primary Language Turkish
Subjects Medical Physiology
Journal Section Research Articles
Authors

Fatih Ahmet Şenel 0000-0003-1918-7277

Rahime Rana Saygın 0000-0001-8755-7781

Mustafa Saygın 0000-0003-4925-3503

Önder Öztürk 0000-0001-8570-2172

Publication Date June 1, 2021
Submission Date March 11, 2021
Acceptance Date April 3, 2021
Published in Issue Year 2021 Volume: 2 Issue: 1

Cite

Vancouver Şenel FA, Saygın RR, Saygın M, Öztürk Ö. Makine Öğrenmesi Algoritmaları Kullanılarak Vücut Analizi ile Uyku Apnesi Teşhisi. SD. 2021;2(1):6-10.