Activity-Integrated Hidden Markov Model to Predict Calving Time
Abstract
:Simple Summary
Abstract
1. Introduction
2. Proposed Dairy Cow Calving Time Prediction
2.1. Activity Observation
2.2. Data Summarization
2.3. Integrated Hidden Markov Model for Calving Time Prediction
- If the observed value for posture changes is less than the sample mean (μ) then the activity is at a low level, and denoted by L.
- If the observed value for posture changes lies between (μ) and (μ +σ) then the activity is at a medium level, and denoted by M.
- If the observed value for posture changes is greater than (μ +σ) then the activity is at a high level, and denoted by H.
- where a11 = transition probability from non-calving state to non-calving state,
- a12 = transition probability from non-calving state to calving state,
- a21 = transition probability from calving state to non-calving state,
- a22 = transition probability from calving state to calving state.
- where, Π1 = probability of non-calving,
- Π2 = probability of calving
2.4. Methodology Implementation Procedure
- L: n < µ,
- M: µ < n < µ + σ,
- H: n > µ + σ
3. Experimental Results
4. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Mee, J.F. Why Do So Many Calves Die on Modern Dairy Farms and What Can We Do about Calf Welfare in the Future? Animal 2013, 3, 1036–1057. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Streyl, D.; Sauter-Louis, C.; Braunert, A.; Lange, D.; Weber, F.; Zerbe, H. Establishment of a standard operating procedure for predicting the time of calving in cattle. J. Vet. Sci. 2011, 12, 177–185. [Google Scholar] [CrossRef] [PubMed]
- Mee, J.F.; Berry, D.; Cromie, A. Risk factors for calving assistance and dystocia in pasture-based Holstein–Friesian heifers and cows in Ireland. Vet. J. 2011, 187, 189–194. [Google Scholar] [CrossRef] [PubMed]
- Jensen, M.B. Behavior around the time of calving in dairy cows. Appl. Anim. Behav. Sci. 2012, 139, 195–202. [Google Scholar] [CrossRef]
- Mee, J.; English, L. Prediction of Onset of Calving from Tail Elevation—Preliminary Evaluation of a Novel Biosensor; Animal & Grassland Research and Innovation Centre, Moorepark, Fermoy Co.: Cork, Ireland, July 2019; pp. 156–157. [Google Scholar]
- Maltz, E.; Antler, A. A practical way to detect approaching calving of the dairy cow by a behaviour sensor. In Proceedings Precision Livestock Farming; Wageningen Academic Publishers: Skiathos, Greece, 2007; pp. 141–146. [Google Scholar]
- Borchers, M.R.; Chang, Y.M.; Proudfoot, K.L.; Wadsworth, B.A.; Stone, A.E.; Bewley, J.M. Machine-learning-based calving prediction from activity, lying, and ruminating behaviors in dairy cattle. J. Dairy Sci. 2017, 100, 5664–5674. [Google Scholar] [CrossRef]
- Fadul, M.; Bogdahn, C.; Alsaaod, M.; Hüsler, J.; Starke, A.; Steiner, A.; Hirsbrunner, G. Prediction of calving time in dairy cattle. Anim. Reprod. Sci. 2017, 187, 37–46. [Google Scholar] [CrossRef]
- Lammoglia, M.A.; Bellows, R.A.; Short, R.E.; Bellows, S.E.; Bighorn, E.G.; Stevenson, J.S.; Randel, R.D. Body temperature and endocrine interactions before and after calving in beef cows. J. Anim. Sci. 1997, 75, 2526–2534. [Google Scholar] [CrossRef] [Green Version]
- Miedema, H.M.; Cockram, M.S.; Dwyer, C.M.; Macrae, A.I. Changes in the behavior of dairy cows during the 24 h before normal calving compared with behavior during late pregnancy. Appl. Anim. Behav. Sci. 2011, 131, 8–14. [Google Scholar] [CrossRef]
- Ouellet, V.; Vasseur, E.; Heuwieser, W.; Burfeind, O.; Maldague, X.; Charbonneau, É. Evaluation of calving indicators measured by automated monitoring devices to predict the onset of calving in Holstein dairy cows. J. Dairy Sci. 2016, 99, 1539–1548. [Google Scholar] [CrossRef] [Green Version]
- Higaki, S.; Koyama, K.; Sasaki, Y.; Abe, K.; Honkawa, K.; Horii, Y.; Minamino, T.; Mikurino, Y.; Okada, H.; Miwakeichi, F.; et al. Technical note: Calving prediction in dairy cattle based on continuous measurements of ventral tail base skin temperature using supervised machine learning. J. Dairy Sci. 2020, 103, 8535–8540. [Google Scholar] [CrossRef]
- Saint-Dizier, M.; Chastant-Maillard, S. Methods and on-farm devices to predict calving time in cattle. Vet. J. 2015, 205, 349–356. [Google Scholar] [CrossRef] [PubMed]
- Rutten, C.J.; Kamphuis, C.; Hogeveen, H.; Huijps, K.; Nielen, M.; Steeneveld, W. Sensor data on cow activity, rumination, and ear temperature improve prediction of the start of calving in dairy cows. Comput. Electron. Agric. 2017, 132, 108–118. [Google Scholar] [CrossRef]
- Proudfoot, K.; Jensen, M.; Weary, D.; Von Keyserlingk, M.A. Dairy cows seek isolation at calving and when ill. J. Dairy Sci. 2014, 97, 2731–2739. [Google Scholar] [CrossRef] [PubMed]
- Schirmann, K.; Chapinal, N.; Weary, D.; Vickers, L.; Von Keyserlingk, M. Short communication: Rumination and feeding behavior before and after calving in dairy cows. J. Dairy Sci. 2013, 96, 7088–7092. [Google Scholar] [CrossRef] [PubMed]
- Huzzey, J.; Von Keyserlingk, M.; Weary, D. Changes in Feeding, Drinking, and Standing Behavior of Dairy Cows During the Transition Period. J. Dairy Sci. 2005, 88, 2454–2461. [Google Scholar] [CrossRef] [Green Version]
- Rice, C.A.; Eberhart, N.L.; Krawczel, P.D. Prepartum lying behavior of holstein dairy cows housed on pasture through parturition. Animals 2017, 7, 32. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Mee, J. Managing the dairy cow at calving time. Vet. Clin. N. Am. Food Anim. Pract. 2004, 20, 521–546. [Google Scholar] [CrossRef]
- Cangar, O.; Leroy, T.; Guarino, M.; Vranken, E.; Fallon, R.; Lenehan, J.; Mee, J.; Berckmans, D. Automatic real-time monitoring of locomotion and posture behavior of pregnant cows prior to calving using online image analysis. Comput. Electron. Agric. 2008, 64, 53–60. [Google Scholar] [CrossRef]
- Ferreira, A.G.; Henrique, D.S.; Vieira, R.A.; Maeda, E.M.; Valotto, A.A. Fitting mathematical models to lactation curves from holstein cows in the southwestern region of the state of Parana, Brazil. Anais Acad. Bras. Ciências 2015, 87, 503–517. [Google Scholar] [CrossRef] [Green Version]
- Burfeind, O.; Suthar, V.; Voigtsberger, R.; Bonk, S.; Heuwieser, W. Validity of prepartum changes in vaginal and rectal temperature to predict calving in dairy cows. J. Dairy Sci. 2011, 94, 5053–5061. [Google Scholar] [CrossRef] [Green Version]
- Fenlon, C.; O’Grady, L.; Mee, J.F.; Butler, S.T.; Doherty, M.L.; Dunnion, J. A comparison of 4 predictive models of calving assistance and difficulty in dairy heifers and cows. J. Dairy Sci. 2017, 100, 9746–9758. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Marchesi, G.; Leonardi, S.; Tangorra, F.M.; Calcante, A.; Beretta, E.; Pofcher, E.; Lazzari, M. Evaluation of an electronic sys-tem for automatic calving detection on a dairy farm. Anim. Product. Sci. 2013, 53, 1112–1114. [Google Scholar] [CrossRef]
- Sumi, K.; Zin, T.T.; Kobayashi, I.; Horii, Y. A Study on Cow Monitoring System for Calving Process. In Proceedings of the IEEE 6th Global Conference on Consumer Electronics (GCCE 2017), Nagoya, Japan, 24–27 October 2017; pp. 379–380. [Google Scholar] [CrossRef]
- Sumi, K.; Zin, T.T.; Kobayashi, I.; Horii, Y. Framework of Cow Calving Monitoring System Using a Single Depth Camera. In Proceedings of the 2018 International Conference on Image and Vision Computing New Zealand (IVCNZ), Auckland, New Zealand, 19–21 November 2018. [Google Scholar] [CrossRef]
- Zin, T.T.; Mar, C.C.; Sumi, K. Background Modelling Using Temporal Average Filter and Running Gaussian Average. In Proceedings of the International Workshop on Frontiers of Computer Vision (IW-FCV), Ibusuki, Japan, 20–22 February 2020. [Google Scholar]
- Zin, T.T.; Maung, S.Z.M.; Tin, P.; Horii, Y. Feature Detection and Classification of Cow Motion for Predicting Calving time. In Proceedings of the 2020 IEEE 9th Global Conference on Consumer Electronics (GCCE), Kobe, Japan, 13–16 October 2020; pp. 305–306. [Google Scholar]
ID | Recording Started Date and Time | Calving Date and Time | ||
---|---|---|---|---|
(mm/dd/yy) | (h/min/s) | (mm/dd/yy) | (h/min/s) | |
1 | 11.29.2017 | (00:00:00) | 12.02.2017 | (12:32:50) |
2 | 11.26.2017 | (17:10:00) | 11.29.2017 | (17:10:00) |
3 | 11.26.2017 | (19:35:00) | 11.29.2017 | (19:35:00) |
4 | 12.04.2017 | (10:05:00) | 12.07.2017 | (10:06:35) |
5 | 11.30.2017 | (15:10:00) | 12.03.2017 | (15:10:00) |
6 | 11.30.2017 | (21:15:00) | 12.03.2017 | (21:15:00) |
7 | 12.04.2017 | (10:10:00) | 12.07.2017 | (10:13:00) |
8 | 12.04.2017 | (16:05:00) | 12.07.2017 | (16:09:40) |
9 | 12.04.2017 | (14:00:00) | 12.06.2017 | (20:10:00) |
10 | 12.11.2017 | (06:00:00) | 12.14.2017 | (05:58:50) |
11 | 12.06.2017 | (10:00:00) | 12.08.2017 | (03:25:00) |
12 | 12.12.2017 | (04:50:00) | 12.15.2017 | (04:53:09) |
13 | 12.07.2017 | (17:20:00) | 12.10.2017 | (17:20:00) |
14 | 12.13.2017 | (21:00:00) | 12.16.2017 | (21:03:29) |
15 | 12.16.2017 | (21:55:00) | 12.19.2017 | (21:55:00) |
16 | 12.14.2017 | (17:15:00) | 12.17.2017 | (17:19:20) |
17 | 12.17.2017 | (06:10:00) | 12.20.2017 | (06:10:00) |
18 | 12.14.2017 | (16:15:00) | 12.17.2017 | (16:17:12) |
19 | 12.17.2017 | (09:50:50) | 12.20.2017 | (09:50:00) |
20 | 12.15.2017 | (01:25:00) | 12.18.2017 | (01:25:21) |
21 | 12.17.2017 | (12:15:00) | 12.20.2017 | (12:15:00) |
22 | 12.09.2017 | (17:15:00) | 12.12.2017 | (17:15:00) |
23 | 12.01.2017 | (10:25:00) | 12.04.2017 | (10:25:18) |
24 | 12.03.2017 | (02:40:00) | 12.06.2017 | (02:41:22) |
25 | 11.29.2017 | (00:30:00) | 12.02.2017 | (00:30:00) |
Variable Name | Unit Measure | Descriptions |
---|---|---|
Lying Time | Minute/Interval | The time in minutes that a cow has spent lying with in a predefined interval |
Standing Time | Minute/Interval | The time in minutes that a cow has spent standing with in a predefined interval |
Lying Bouts | Number/Interval | The number of transitions from standing to lying during the interval |
Standing Bouts | Number/Interval | The number of transitions from lying to standing during the interval |
Cow ID 1 | ||||||||
---|---|---|---|---|---|---|---|---|
Time | Posture Changes | Activity Level | Time | Posture Change | Activity Level | Time | Posture Change | Activity Level |
−72 | 2 | Mid | −48 | 2 | Mid | −24 | 3 | Mid |
−71 | 2 | Mid | −47 | 4 | High | −23 | 3 | Mid |
−70 | 5 | High | −46 | 2 | Mid | −22 | 4 | High |
−69 | 6 | High | −45 | 3 | Mid | −21 | 2 | Mid |
−68 | 1 | Low | −44 | 1 | Low | −20 | 4 | High |
−67 | 2 | Mid | −43 | 2 | Mid | −19 | 2 | Mid |
−66 | 1 | Low | −42 | 1 | Low | −18 | 2 | Mid |
−65 | 0 | Low | −41 | 0 | Low | −17 | 0 | Low |
−64 | 0 | Low | −40 | 1 | Low | −16 | 0 | Low |
−63 | 2 | Mid | −39 | 2 | Mid | −15 | 0 | Low |
−62 | 2 | Mid | −38 | 2 | Mid | −14 | 0 | Low |
−61 | 2 | Mid | −37 | 1 | Low | −13 | 4 | High |
−60 | 2 | Mid | −36 | 4 | High | −12 | 2 | Mid |
−59 | 1 | Low | −35 | 1 | Low | −11 | 0 | Low |
−58 | 2 | Mid | −34 | 2 | Mid | −10 | 1 | Low |
−57 | 2 | Mid | −33 | 1 | Low | −9 | 1 | Low |
−56 | 1 | Low | −32 | 0 | Low | −8 | 2 | Mid |
−55 | 0 | Low | −31 | 0 | Low | −7 | 2 | Mid |
−54 | 0 | Low | −30 | 0 | Low | −6 | 2 | Mid |
−53 | 0 | Low | −29 | 1 | Low | −5 | 6 | High |
−52 | 1 | Low | −28 | 4 | High | −4 | 2 | Mid |
−51 | 1 | Low | −27 | 2 | Mid | −3 | 2 | Mid |
−50 | 2 | Mid | −26 | 3 | Mid | −2 | 5 | High |
−49 | 1 | Low | −25 | 0 | Low | −1 | 13 | High |
Sample Mean µ of posture change frequency | 1.931 |
Standard Deviation σ of posture change frequency | 1.967 |
(µ + σ) | 3.897 |
(µ + 2σ) | 5.864 |
Actual Time | 3 h Interval 1–24 | Activity | States | Actual Time | 3 h Interval 1–24 | Activity | States |
---|---|---|---|---|---|---|---|
−72 | −24 | L | NC | −36 | −12 | L | NC |
−69 | −23 | L | NC | −33 | −11 | L | NC |
−66 | −22 | L | NC | −30 | −10 | L | NC |
−63 | −21 | L | NC | −27 | −9 | L | NC |
−60 | −20 | L | NC | −24 | −8 | M | NC |
−57 | −19 | L | NC | −21 | −7 | L | NC |
−54 | −18 | L | NC | −18 | −6 | L | NC |
−51 | −17 | L | NC | −15 | −5 | L | NC |
−48 | −16 | L | NC | −12 | −4 | L | NC |
−45 | −15 | L | NC | −9 | −3 | L | NC |
−42 | −14 | L | NC | −6 | −2 | M | NC |
−39 | −13 | L | NC | −3 | −1 | H | C |
No. | Sensitivity (%) | Precision (%) |
---|---|---|
1 | 91.30 | 95.45 |
2 | 86.96 | 95.24 |
3 | 90.63 | 87.88 |
4 | 95.83 | 100 |
5 | 91.67 | 100 |
6 | 90.48 | 86.36 |
7 | 91.67 | 100 |
8 | 91.67 | 100 |
9 | 85 | 80.95 |
10 | 95.24 | 86.96 |
Average | 91.05 | 93.28 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Sumi, K.; Maw, S.Z.; Zin, T.T.; Tin, P.; Kobayashi, I.; Horii, Y. Activity-Integrated Hidden Markov Model to Predict Calving Time. Animals 2021, 11, 385. https://doi.org/10.3390/ani11020385
Sumi K, Maw SZ, Zin TT, Tin P, Kobayashi I, Horii Y. Activity-Integrated Hidden Markov Model to Predict Calving Time. Animals. 2021; 11(2):385. https://doi.org/10.3390/ani11020385
Chicago/Turabian StyleSumi, Kosuke, Swe Zar Maw, Thi Thi Zin, Pyke Tin, Ikuo Kobayashi, and Yoichiro Horii. 2021. "Activity-Integrated Hidden Markov Model to Predict Calving Time" Animals 11, no. 2: 385. https://doi.org/10.3390/ani11020385
APA StyleSumi, K., Maw, S. Z., Zin, T. T., Tin, P., Kobayashi, I., & Horii, Y. (2021). Activity-Integrated Hidden Markov Model to Predict Calving Time. Animals, 11(2), 385. https://doi.org/10.3390/ani11020385