Analisis Kualitas Sperma Menggunakan Adaptive Object Tracking untuk Perolehan Data Motilitas dan Kecepatan Sperma pada Rekaman Mikroskopis Digital

Aufa, Izzaz (2024) Analisis Kualitas Sperma Menggunakan Adaptive Object Tracking untuk Perolehan Data Motilitas dan Kecepatan Sperma pada Rekaman Mikroskopis Digital. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5023201026-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (2MB) | Request a copy

Abstract

Kemandulan adalah gangguan pada sistem reproduksi yang ditandai oleh ketidakmampuan pasangan untuk mencapai kehamilan setelah berhubungan seksual tanpa perlindungan atau kontrasepsi selama 12 bulan. Di Indonesia, sekitar 10-15% pasangan mengalami kemandulan, artinya sekitar 4-6 juta pasangan dari total 39,8 juta pasangan usia subur namun banyak dari mereka masih belum menyadarinya. Selama ini kemandulan diperiksa melalui pengamatan hasil mikroskopis dari sperma subjek, hal ini menyebabkan adanya unsur subjektifitas dalam diagnosa adanya kemandulan pada pasien terlebih lagi adanya peluang terjadinya human error pada hasil analisanya. Pada penelitian tugas akhir ini, diajukan sistem untuk memperoleh data pola gerak dan motilitas sperma dari rekaman mikroskop menggunakan metode ADTARI (Adaptive Object Tracking Algorithm) sebagai bentuk CASA (Computer Assisted Sperm Analysis). Video rekaman dipisah menjadi frame-frame (30 fps), yang kemudian diolah menggunakan geometric mean filter untuk menghilangkan noise, dengan PSNR sebesar 35,94 dB dan MSE sebesar 13,11. Selanjutnya, segmentasi sperma dilakukan menggunakan model U-Net yang dilatih dengan 420 frame dan diuji dengan 20 frame. Model dengan 75 epoch memberikan hasil terbaik, dengan dice coefficient sebesar 89,9% dan IoU sebesar 90,4%. Segmentasi ini membantu menentukan titik centroid kepala sperma menggunakan metode contouring, dengan akurasi 83%. Data pola gerak dan motilitas sperma diperoleh melalui adaptive object tracking algorithm, menghasilkan pola gerak individu dan histogram pergerakan sperma. Data ini digunakan untuk menghitung VAP dan VSL, yang penting untuk menilai kesehatan sperma menurut WHO. Metode ini efisien karena memproses frame-frame yang telah dipisahkan dan diolah sebelumnya, bukan langsung dari video.
========================================================================================================================
Infertility is a disorder of the reproductive system characterized by a couple's inability to achieve pregnancy after having unprotected sexual intercourse for 12 months. In Indonesia, around 10-15% of couples experience infertility, which translates to approximately 4-6 million couples out of a total of 39.8 million couples of reproductive age, many of whom are still unaware of their condition. Traditionally, infertility is examined through microscopic observation of the subject's sperm, which introduces subjectivity in diagnosing infertility and the potential for human error in the analysis results. In this final project research, a system is proposed to obtain sperm motility and movement pattern data from microscope recordings using the ADTARI (Adaptive Object Tracking Algorithm) method as a form of CASA (Computer Assisted Sperm Analysis). The video recordings are split into frames (30 fps), which are then processed using a geometric mean filter to remove noise, resulting in a PSNR of 35.94 dB and an MSE of 13.11. Next, sperm segmentation is performed using a U-Net model trained with 420 frames and tested with 20 frames. The model with 75 epochs provided the best results, with a dice coefficient of 89.9% and an IoU of 90.4%. This segmentation helps determine the centroid of the sperm head using the contouring method, with an accuracy of 83%. Sperm motility and movement pattern data are obtained through the adaptive object tracking algorithm, producing individual movement patterns and a movement histogram of the sperm. This data is used to calculate VAP (Average Path Velocity) and VSL (Straight Line Velocity), which are important for assessing sperm health according to WHO standards. This method is efficient because it processes pre-separated and pre-processed frames rather than the video directly.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Kemandulan, Computer Assisted Sperm Analysis (CASA), Motilitas ======================================================================================================================== ======================================================================================================================== Infertility, Computer Assisted Sperm Analysis (CASA), Sperm Motility, Adaptive Object Tracking Algorithm Sperma, Adaptive Object Tracking Algorithm ======================================================================================================================== Infertility, Computer Assisted Sperm Analysis (CASA), Sperm Motility, Adaptive Object Tracking Algorithm
Subjects:	R Medicine > R Medicine (General) > R858 Deep Learning T Technology > T Technology (General) > T59.7 Human-machine systems. T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TA Engineering (General). Civil engineering (General) > TA174 Computer-aided design.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Izzaz Aufa
Date Deposited:	12 Aug 2024 07:44
Last Modified:	12 Aug 2024 07:44
URI:	http://repository.its.ac.id/id/eprint/112720

Actions (login required)

View Item