View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Physics
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Physics
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Penggunaan Random Forest dalam Analisis Pengukuran Model Kadar Hemoglobin dengan Spektrofotometri

      Thumbnail
      View/Open
      Cover (402.7Kb)
      Fullteks (1.808Mb)
      Lampiran (319.5Kb)
      Date
      2023
      Author
      Assilmi, Fadmiar Nibras
      Zuhri, Mahfuddin
      Irzaman
      Metadata
      Show full item record
      Abstract
      Kadar hemoglobin merupakan parameter klinis yang penting untuk menilai anemia dalam kondisi kronis dan akut. Pengukuran kadar hemoglobin darah saat ini kebanyakan menggunakan metode invasif yang kurang akurat dan efisien. Proses pengukuran dapat dilakukan secara noninvasif yang dianggap lebih akurat, lebih cepat, dan tidak menyakiti pasien. Penelitian ini bertujuan menguji performa metode Random Forest untuk menganalisis pengukuran kadar hemoglobin darah dengan metode pengujian akurasi dan diagnostik berupa Root Mean Square Error (RMSE), sensitivitas, spesifisitas, diagnosis akurasi, Number Needed to Diagnose (NND). Hasil penelitian ini disimpulkan bahwa kandidat panjang gelombang yang memiliki nilai akurasi paling tinggi yaitu pada rentang panjang gelombang 365 nm dengan nilai RMSE 0,56 dan 385 nm sampai 395 nm dengan nilai RMSE 0,57. Hasil analisis sensitivitas, spesifisitas, diagnosis akurasi, dan number needed to diagnose menunjukkan ketiga kandidat panjang gelombang memiliki nilai yang sama dan memenuhi kriteria untuk masing-masing parameter dengan nilai 100%, 100%, 100%, dan 1 kali pengulangan. Nilai korelasi Pearson juga mencapai angka yang tinggi bernilai 0,97.
       
      Hemoglobin level is an important clinical parameter for assessing anemia in chronic and acute conditions. Current measurement of blood hemoglobin levels mostly uses invasive methods that are less accurate and efficient. The measurement process can be carried out non-invasively which is considered more accurate, faster and does not hurt the patient. This study aims to test the performance of the Random Forest method to analyze measurements of blood hemoglobin levels with accuracy and diagnostic testing methods in the form of Root Mean Square Error (RMSE), sensitivity, specificity, diagnosis accuracy, Number Needed to Diagnose (NND). The results of this study concluded that the candidate wavelengths that have the highest accuracy values are in the wavelength range of 365 nm with an RMSE value of 0.56 and 385 nm to 395 with an RMSE value of 0.57. The results of the analysis of sensitivity, specificity, accuracy diagnosis, and number needed for diagnosis show that the three wavelength candidates have the same value and meet the criteria for each parameter with a value of 100%, 100%, 100%, and 1 repetition. The Pearson correlation value also reaches a high value of 0.97.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/124446
      Collections
      • UT - Physics [1230]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository