Validation of an Automated Body Condition Scoring System Using 3D Imaging
Body condition scores (BCS) measure a cow’s fat reserves and is important for management and research. Manual BCS assessment is subjective, time-consuming, and requires trained personnel. The BodyMat F (BMF, Ingenera SA, Cureglia, Switzerland) is an automated body condition scoring system using a 3D...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-06-01
|
| Series: | Agriculture |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-0472/10/6/246 |
| _version_ | 1797564065400225792 |
|---|---|
| author | Niall O’ Leary Lorenzo Leso Frank Buckley Jonathon Kenneally Diarmuid McSweeney Laurence Shalloo |
| author_facet | Niall O’ Leary Lorenzo Leso Frank Buckley Jonathon Kenneally Diarmuid McSweeney Laurence Shalloo |
| author_sort | Niall O’ Leary |
| collection | DOAJ |
| description | Body condition scores (BCS) measure a cow’s fat reserves and is important for management and research. Manual BCS assessment is subjective, time-consuming, and requires trained personnel. The BodyMat F (BMF, Ingenera SA, Cureglia, Switzerland) is an automated body condition scoring system using a 3D sensor to estimate BCS. This study assesses the BMF. One hundred and three Holstein Friesian cows were assessed by the BMF and two assessors throughout a lactation. The BMF output is in the 0–5 scale commonly used in France. We develop and report the first equation to convert these scores to the 1–5 scale used by the assessors in Ireland in this study ((0–5 scale × 0.38) + 1.67 → 1–5 scale). Inter-assessor agreement as measured by Lin’s concordance of correlation was 0.67. BMF agreement with the mean of the two assessors was the same as between assessors (0.67). However, agreement was lower for extreme values, particularly in over-conditioned cows where the BMF underestimated BCS relative to the mean of the two human observers. The BMF outperformed human assessors in terms of reproducibility and thus is likely to be especially useful in research contexts. This is the second independent validation of a commercially marketed body condition scoring system as far as the authors are aware. Comparing the results here with the published evaluation of the other system, we conclude that the BMF performed as well or better. |
| first_indexed | 2024-03-10T18:52:09Z |
| format | Article |
| id | doaj.art-3d0260b8ff7342579558c3e1572096d1 |
| institution | Directory Open Access Journal |
| issn | 2077-0472 |
| language | English |
| last_indexed | 2024-03-10T18:52:09Z |
| publishDate | 2020-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agriculture |
| spelling | doaj.art-3d0260b8ff7342579558c3e1572096d12023-11-20T05:01:47ZengMDPI AGAgriculture2077-04722020-06-0110624610.3390/agriculture10060246Validation of an Automated Body Condition Scoring System Using 3D ImagingNiall O’ Leary0Lorenzo Leso1Frank Buckley2Jonathon Kenneally3Diarmuid McSweeney4Laurence Shalloo5Hincks Centre For Entrepreneurial Excellence, Cork Institute of Technology, Cork, T12 P928 Co. Cork, IrelandDepartment of Agricultural, Food and Forestry Systems, University of Florence, 50145 Firenze, ItalyMoorepark Animal & Grassland Research and Innovation Centre, Teagasc, Fermoy, P61 C997 Co. Cork, IrelandMoorepark Animal & Grassland Research and Innovation Centre, Teagasc, Fermoy, P61 C997 Co. Cork, IrelandUnit 2, True North Technologies, Shannon Business Centre, Shannon, V14 YT99 Co. Clare, IrelandMoorepark Animal & Grassland Research and Innovation Centre, Teagasc, Fermoy, P61 C997 Co. Cork, IrelandBody condition scores (BCS) measure a cow’s fat reserves and is important for management and research. Manual BCS assessment is subjective, time-consuming, and requires trained personnel. The BodyMat F (BMF, Ingenera SA, Cureglia, Switzerland) is an automated body condition scoring system using a 3D sensor to estimate BCS. This study assesses the BMF. One hundred and three Holstein Friesian cows were assessed by the BMF and two assessors throughout a lactation. The BMF output is in the 0–5 scale commonly used in France. We develop and report the first equation to convert these scores to the 1–5 scale used by the assessors in Ireland in this study ((0–5 scale × 0.38) + 1.67 → 1–5 scale). Inter-assessor agreement as measured by Lin’s concordance of correlation was 0.67. BMF agreement with the mean of the two assessors was the same as between assessors (0.67). However, agreement was lower for extreme values, particularly in over-conditioned cows where the BMF underestimated BCS relative to the mean of the two human observers. The BMF outperformed human assessors in terms of reproducibility and thus is likely to be especially useful in research contexts. This is the second independent validation of a commercially marketed body condition scoring system as far as the authors are aware. Comparing the results here with the published evaluation of the other system, we conclude that the BMF performed as well or better.https://www.mdpi.com/2077-0472/10/6/246Body condition scorecowsautomatedvalidationprecision technology |
| spellingShingle | Niall O’ Leary Lorenzo Leso Frank Buckley Jonathon Kenneally Diarmuid McSweeney Laurence Shalloo Validation of an Automated Body Condition Scoring System Using 3D Imaging Agriculture Body condition score cows automated validation precision technology |
| title | Validation of an Automated Body Condition Scoring System Using 3D Imaging |
| title_full | Validation of an Automated Body Condition Scoring System Using 3D Imaging |
| title_fullStr | Validation of an Automated Body Condition Scoring System Using 3D Imaging |
| title_full_unstemmed | Validation of an Automated Body Condition Scoring System Using 3D Imaging |
| title_short | Validation of an Automated Body Condition Scoring System Using 3D Imaging |
| title_sort | validation of an automated body condition scoring system using 3d imaging |
| topic | Body condition score cows automated validation precision technology |
| url | https://www.mdpi.com/2077-0472/10/6/246 |
| work_keys_str_mv | AT nialloleary validationofanautomatedbodyconditionscoringsystemusing3dimaging AT lorenzoleso validationofanautomatedbodyconditionscoringsystemusing3dimaging AT frankbuckley validationofanautomatedbodyconditionscoringsystemusing3dimaging AT jonathonkenneally validationofanautomatedbodyconditionscoringsystemusing3dimaging AT diarmuidmcsweeney validationofanautomatedbodyconditionscoringsystemusing3dimaging AT laurenceshalloo validationofanautomatedbodyconditionscoringsystemusing3dimaging |