Fast Spot Locating for Low-Density DNA Microarray
Low-density DNA microarrays are crucial in molecular diagnostics due to their cost-effectiveness and high sensitivity. However, reliable spot localization remains challenging due to positional variations and image artifacts. Traditional intensity-based methods often struggle with weak fluorescence s...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/7/2135 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850188360530788352 |
|---|---|
| author | MinGin Kim Jongwon Kim Sun-Hee Kim Jong-Dae Kim |
| author_facet | MinGin Kim Jongwon Kim Sun-Hee Kim Jong-Dae Kim |
| author_sort | MinGin Kim |
| collection | DOAJ |
| description | Low-density DNA microarrays are crucial in molecular diagnostics due to their cost-effectiveness and high sensitivity. However, reliable spot localization remains challenging due to positional variations and image artifacts. Traditional intensity-based methods often struggle with weak fluorescence signals. To address this, we propose a rapid spot localization method that combines template matching with point pattern matching, enhanced through vectorized programming and square (box) templates. Vectorized programming accelerated the most time-consuming calculation by 82 times on a PC and was 6000 times faster on a Raspberry Pi compared to a for-loop implementation. While this improvement applies to the vectorized square calculation alone, substantial performance gains were still achieved in the overall process. Additionally, replacing circular templates with square templates resulted in a fourfold reduction in processing time without compromising detection performance. The proposed method effectively reduces computational overhead, making it suitable for high-throughput and resource-constrained applications. The method was validated using HPV genotyping images from commercial DNA microarrays, demonstrating its practical applicability and robust performance in clinical settings. |
| format | Article |
| id | doaj-art-644c14ea07a64ad09b93802a6eb01e7f |
| institution | OA Journals |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-644c14ea07a64ad09b93802a6eb01e7f2025-08-20T02:15:54ZengMDPI AGSensors1424-82202025-03-01257213510.3390/s25072135Fast Spot Locating for Low-Density DNA MicroarrayMinGin Kim0Jongwon Kim1Sun-Hee Kim2Jong-Dae Kim3Thermo Fisher Scientific, South San Francisco, CA 94080, USABiomedux, Suwon-si 16226, Republic of KoreaDepartment of Fashion Industry, Incheon National University, Incheon 22012, Republic of KoreaSchool of Software, Hallym University, Chuncheon-si 24252, Republic of KoreaLow-density DNA microarrays are crucial in molecular diagnostics due to their cost-effectiveness and high sensitivity. However, reliable spot localization remains challenging due to positional variations and image artifacts. Traditional intensity-based methods often struggle with weak fluorescence signals. To address this, we propose a rapid spot localization method that combines template matching with point pattern matching, enhanced through vectorized programming and square (box) templates. Vectorized programming accelerated the most time-consuming calculation by 82 times on a PC and was 6000 times faster on a Raspberry Pi compared to a for-loop implementation. While this improvement applies to the vectorized square calculation alone, substantial performance gains were still achieved in the overall process. Additionally, replacing circular templates with square templates resulted in a fourfold reduction in processing time without compromising detection performance. The proposed method effectively reduces computational overhead, making it suitable for high-throughput and resource-constrained applications. The method was validated using HPV genotyping images from commercial DNA microarrays, demonstrating its practical applicability and robust performance in clinical settings.https://www.mdpi.com/1424-8220/25/7/2135low-density DNA microarrayspot localizationvectorized programmingsquare template matchingHPV genotyping |
| spellingShingle | MinGin Kim Jongwon Kim Sun-Hee Kim Jong-Dae Kim Fast Spot Locating for Low-Density DNA Microarray Sensors low-density DNA microarray spot localization vectorized programming square template matching HPV genotyping |
| title | Fast Spot Locating for Low-Density DNA Microarray |
| title_full | Fast Spot Locating for Low-Density DNA Microarray |
| title_fullStr | Fast Spot Locating for Low-Density DNA Microarray |
| title_full_unstemmed | Fast Spot Locating for Low-Density DNA Microarray |
| title_short | Fast Spot Locating for Low-Density DNA Microarray |
| title_sort | fast spot locating for low density dna microarray |
| topic | low-density DNA microarray spot localization vectorized programming square template matching HPV genotyping |
| url | https://www.mdpi.com/1424-8220/25/7/2135 |
| work_keys_str_mv | AT minginkim fastspotlocatingforlowdensitydnamicroarray AT jongwonkim fastspotlocatingforlowdensitydnamicroarray AT sunheekim fastspotlocatingforlowdensitydnamicroarray AT jongdaekim fastspotlocatingforlowdensitydnamicroarray |