Letícia Marra Campos

Letícia Marra Campos

Auto-generated profile, based on publicly available information

Letícia Marra Campos

Department of Animal Sciences, Virginia Tech, Blacksburg, VA 24060

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 October 2022 in Translational Animal Science
Verified
Journal Article
Open access
PDF file
RKRobert Kadlec
Robert Kadlec
SISam Indest
Sam Indest
KCKayla Castro
Kayla Castro
SWShayan Waqar
Shayan Waqar
LCLeticia M Campos
Leticia M Campos
GMGota Morota
Gota Morota
Animal dimensions are essential indicators for monitoring their growth rate, diet efficiency, and health status. A computer vision system is a recently emerging precision livestock farming technology that overcomes the previously unresolved challenges pertaining to labor and cost. Depth sensor cameras can be used to estimate the depth or height of an animal, in addition to two-dimensional information. Collecting top-view depth images is common in evaluating body mass or conformational traits in livestock species. However, in the depth image data acquisition process, manual interventions are involved in controlling a camera from a laptop or where detailed steps for automated data collection are not documented. Furthermore, open-source image data acquisition implementations are rarely available. The objective of this study was to 1) investigate the utility of automated top-view dairy cow depth data collection methods using picture- and video-based methods, 2) evaluate the performance of an infrared cut lens, 3) and make the source code available. Both methods can automatically perform animal detection, trigger recording, capture depth data, and terminate recording for individual animals. The picture-based method takes only a predetermined number of images whereas the video-based method uses a sequence of frames as a video. For the picture-based method, we evaluated 3- and 10-picture approaches. The depth sensor camera was mounted 2.75 m above-the-ground over a walk-through scale between the milking parlor and the free-stall barn. A total of 150 Holstein and 100 Jersey cows were evaluated. A pixel location where the depth was monitored was set up as a point of interest. More than 89% of cows were successfully captured using both picture- and video-based methods. The success rates of the picture- and video-based methods further improved to 92% and 98%, respectively, when combined with an infrared cut lens. Although both the picture-based method with 10 pictures and the video-based method yielded accurate results for collecting depth data on cows, the former was more efficient in terms of data storage. The current study demonstrates automated depth data collection frameworks and a Python implementation available to the community, which can help facilitate the deployment of computer vision systems for dairy cows.

5 October 2023 in Animal
Verified
Journal Article
Open access
LCL.M. Campos
L.M. Campos
HRH. Ringer
H. Ringer
MCM. Chung
M. Chung
HAHanigan
Hanigan

1 February 2021 in Journal of Dairy Science
Verified
Journal Article
Closed access
XHX. Huang
X. Huang
PYP.S. Yoder
P.S. Yoder
LCL. Campos
L. Campos
EHE. Huang
E. Huang
MHMark D. Hanigan
Mark D. Hanigan
The objective of this study was to extend a stable isotope-based assessment of AA absorption from rumen-degradable protein (RDP) sources to include determination of essential AA (EAA) availability from microbial protein (MCP). To demonstrate the technique, a study using a 2 × 2 factorial arrangement of treatments applied in a repeated 4 × 4 Latin square design was undertaken. Factors were high and low rumen-degradable protein and high and low starch. Twelve lactating cows were blocked into 3 groups according to days in milk and randomly assigned to the 4 treatment sequences. Each period was 14 d in length with 10 d of adaption followed by 4 d of ruminal infusions of 15N-labeled ammonium sulfate. On the last day of each period, a 13C-labeled AA mixture was infused into the jugular vein over a 6-h period to assess total AA entry. Rumen, blood, urine, and milk samples were collected during the infusions. Ruminal bacteria and blood samples were assessed for AA enrichment. Total plasma AA absorption rates were derived for 6 EAA from plasma 13C AA enrichment. Absorption of 6 EAA from MCP was calculated from total AA absorption based on 15N enrichment in blood and rumen bacteria. Essential AA absorption rates from total protein, MCP, and rumen-undegradable protein were derived with standard errors of the mean of 6, 14, and 14%, respectively. An average of 45% of absorbed EAA were from MCP, which varied among 6 EAA and was interactively affected by starch and RDP in diets. Microbial AA availability measured by isotope dilution method increased with the high RDP diets and was unaffected by starch level, except for Met, which decreased with high starch. Microbial protein outflow, estimated from urinary purine derivatives, increased with RDP and was not significantly affected by starch. This was consistent with measurements from the isotope dilution method. Total AA absorption rates measured from isotope dilution were similar to estimates from CNCPS (v. 6.55), but a lower proportion of absorbed AA was derived from MCP for the former method. Compared with the isotope and CNCPS estimates, the Fleming model underestimated microbial EAA and total EAA availability. An average of 58% of the absorbed EAA was converted into milk, which varied among individual AA and was interactively affected by starch and RDP in diets. The isotope dilution approach is advantageous because it provides estimates of EAA availability for individual EAA from rumen-undegradable protein and MCP directly with fewer errors of measurement than can be achieved with intestinal disappearance methods.

1 November 2020 in Animal Feed Science and Technology
Verified
Journal Article
Closed access
VSVinícius Carneiro de Souza
Vinícius Carneiro de Souza
JMJuliana Duarte Messana
Juliana Duarte Messana
EBErick Darlisson Batista
Erick Darlisson Batista
KAKênia Larissa Gomes Carvalho Alves
Kênia Larissa Gomes Carvalho Alves
Ana Veronica Lino DiasAna Veronica Lino Dias
Ana Veronica Lino Dias
TBTelma Teresinha Berchielli
Telma Teresinha Berchielli

21 March 2022 in Journal of Animal Science
Verified
Journal Article
Closed access
PDF file
LCLetícia M Campos
Letícia M Campos
VSVinicius Carneiro de Souza
Vinicius Carneiro de Souza
YGYury T Granja-Salcedo
Yury T Granja-Salcedo
JMJULIANA DUARTE MESSANA
JULIANA DUARTE MESSANA
JPJacquelyn M Prestegaard-Wilson
Jacquelyn M Prestegaard-Wilson
TBTelma T Berchielli
Telma T Berchielli
This study aimed to evaluate the use of total odd-chain fatty acids (OCFA) as a marker to estimate microbial nitrogen flow (MicN) and calculate the efficiency of microbial nitrogen synthesis (EMNS) in Nellore steers fed high-concentrate diets supplemented with different nitrogen supplements (NS). Ruminally and duodenally cannulated Nellore steers (n = 6; 354 ± 12 kg) were used in a 6 × 6 repeated switchback design balanced for residual effects. Treatments were arranged in a 3 × 3 factorial of three nitrogen (N) supplements (urea plus soybean meal; corn gluten meal; dried distillers’ grains plus solubles) and three microbial markers (OCFA; double-labeled urea, 15N; microbial nucleic acid bases, MNAB). The total mixed ration was composed of fresh chopped sugarcane as the forage source in an 83:17 concentrate: forage ratio (dry matter basis). Linear regression was used to develop predictions of MicN from OCFA using 15N and MNAB as response variables. Microbial N flow was underestimated by the MNAB marker compared to 15N. Neither NS nor their respective interactions with the marker methods (MM) affected MicN or EMNS (P > 0.05). However, MicN was different for 15N and MNAB (P > 0.001 for both treatments). Marker methods affected EMNS in all energetic bases (total digestible carbohydrates P < 0.001; rumen-fermentable carbohydrates P < 0.001; organic matter truly degradable in the rumen P < 0.001). Equations that utilized OCFA as a regressor to predict MicN under different MM resulted in good fits of the data as observed by the coefficient of determination (R2; 15N = 0.78; MNAB = 0.69). Microbial N flow estimated from OCFA was overpredicted (15N by 7.46%; MNAB by 4.30%) compared with observed values. The OCFA model presented a small slope bias when methodological validation was applied (15N = 0.96%; MNAB = 3.90%), ensuring reliability of the proposed alternative method. Based on the conditions of this experiment, OCFA may be a suitable alternative to other methods that quantify MicN under different dietary conditions. Nutritional strategies that maximize microbial nitrogen supply to the small intestine may improve cattle performance. Nevertheless, in vivo quantification generally requires sensitive or expensive methods and often yields highly variable results. In the present work, we investigated the use of duodenal odd-chain fatty acids (OCFA) as an alternative method to predict microbial nitrogen flow (MicN) and calculated its efficiency on different energetic bases under different dietary nitrogen supplements. We utilized total OCFA flow (TOCFAf) to predict MicN by two well-established conventional methods: 1) 15N, considered the gold standard and 2) microbial nucleic acid bases. Models presented a positive relationship between TOCFDf and response variables, and under validation, both demonstrated low estimation bias. Under the conditions of this experiment, OCFA appeared to serve as an alternative marker to quantify ruminal MicN for beef cattle.

6 November 2023 in Journal of Animal Science
Verified
Journal Article
Open access
PDF file
YBYe Bi
Ye Bi
LCLeticia Campos
Leticia Campos
JWJin Wang
Jin Wang
HYHaipeng Yu
Haipeng Yu
MHMark Hanigan
Mark Hanigan
GMGota Morota
Gota Morota
Cow body weight monitoring is critical in farm management because it is associated with the growth, nutritional status, and health of dairy cows. Regularly collecting body weight records often requires human labor or expensive equipment. Recently, a low-cost 3D depth sensor camera-enabled computer vision analysis has been proposed to address the body weight prediction challenges. Cow body weight is a repeated trait; however, past body weight prediction research only used data collected at a single time point. Furthermore, the utility of deep learning on body weight predictions remains unanswered. Therefore, the objectives of this study were to predict cow body weight from repeatedly measured video data and compare the performance of thresholding and Mask R-CNN deep learning approaches. We set up an Intel Realsense D435 camera at the Virginia Tech Kentland farm to collect top-view videos for 10 Holstein and 2 Jersey cows for 28 days, twice per day. Real body weight records were also collected by a walk-over-weighing system simultaneously. A total of 40,405 depth images and depth map files were obtained. We explored three approaches to segment the cow body from the background, including single thresholding, adaptive thresholding, and Mask R-CNN. Four image-derived biometric measurements, length, width, height, and volume, were estimated from segmented images and fitted in ordinary least squares. Two cross-validation designs, forecasting and leave-several-cow-out, were used to evaluate prediction performance. The correlation between image-derived biometric measurements and body weight ranged from 0.84 to 0.95. On average, the Mask-RCNN approach resulted in the best prediction coefficient of determination and mean absolute percentage error of 0.96 and 3.54% in the forecasting cross-validation, respectively. The predictive performance of the single and adaptive thresholding approaches was similar. The Mask-RCNN approach was also the best in the leave-several-cow-out cross-validation, followed by adaptive and single thresholding. The prediction coefficients of determination of Mask-RCNN for leave-one-cow-out, leave-two-cow-out, leave-three-cow-out, leave-four-cow-out were 0.05, 0.69, 0.85, and 0.90, respectively, while mean absolute percentage errors were 4.51%, 4.51%, and 4.59%, 4.70% respectively. Our results suggest predicting cow body weight from depth video data using deep learning is feasible.