Relationship Between Somatic Cell Counts and Mammary Gland Parenchyma Ultrasonography in Buffaloes

Zhang, Xinxin; Ahmad, Muhammad Jamil; An, Zhigao; Niu, Kaifeng; Wang, Wei; Nie, Pei; Gao, Shan; Yang, Liguo

doi:10.3389/fvets.2022.842105

Cited by 4 publications

(4 citation statements)

References 40 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ultrasound scanning of the mammary parenchyma at hour 72 after Staphylococcus infection showed a large range of hyperechoic areas, the lactiferous ducts became quite narrow, and the anechoic blood vessels could not be displayed [ 11 ]. Zhang et al [ 39 ] used computer-assisted technology to quantify the texture of udder ultrasound images and found that there was a positive correlation between the pixel standard deviation of udder ultrasound texture and the SCC level, indicating that, as the mastitis became increasingly severe, the pathological manifestations corresponding to the ultrasound texture became more obvious. The report by Abdullah et al [ 40 ] found that, as mastitis intensifies, the average numerical pixel values and pixel heterogeneity of the udder parenchyma echotexture increased.…”

Section: Resultsmentioning

confidence: 99%

A New Method to Detect Buffalo Mastitis Using Udder Ultrasonography Based on Deep Learning Network

Zhang,

Li,

Zhang

et al. 2024

Animals

Self Cite

View full text Add to dashboard Cite

Mastitis is one of the most predominant diseases with a negative impact on ranch products worldwide. It reduces milk production, damages milk quality, increases treatment costs, and even leads to the premature elimination of animals. In addition, failure to take effective measures in time will lead to widespread disease. The key to reducing the losses caused by mastitis lies in the early detection of the disease. The application of deep learning with powerful feature extraction capability in the medical field is receiving increasing attention. The main purpose of this study was to establish a deep learning network for buffalo quarter-level mastitis detection based on 3054 ultrasound images of udders from 271 buffaloes. Two data sets were generated with thresholds of somatic cell count (SCC) set as 2 × 105 cells/mL and 4 × 105 cells/mL, respectively. The udders with SCCs less than the threshold value were defined as healthy udders, and otherwise as mastitis-stricken udders. A total of 3054 udder ultrasound images were randomly divided into a training set (70%), a validation set (15%), and a test set (15%). We used the EfficientNet_b3 model with powerful learning capabilities in combination with the convolutional block attention module (CBAM) to train the mastitis detection model. To solve the problem of sample category imbalance, the PolyLoss module was used as the loss function. The training set and validation set were used to develop the mastitis detection model, and the test set was used to evaluate the network’s performance. The results showed that, when the SCC threshold was 2 × 105 cells/mL, our established network exhibited an accuracy of 70.02%, a specificity of 77.93%, a sensitivity of 63.11%, and an area under the receiver operating characteristics curve (AUC) of 0.77 on the test set. The classification effect of the model was better when the SCC threshold was 4 × 105 cells/mL than when the SCC threshold was 2 × 105 cells/mL. Therefore, when SCC ≥ 4 × 105 cells/mL was defined as mastitis, our established deep neural network was determined as the most suitable model for farm on-site mastitis detection, and this network model exhibited an accuracy of 75.93%, a specificity of 80.23%, a sensitivity of 70.35%, and AUC 0.83 on the test set. This study established a 1/4 level mastitis detection model which provides a theoretical basis for mastitis detection in buffaloes mostly raised by small farmers lacking mastitis diagnostic conditions in developing countries.

show abstract

Section: Resultsmentioning

confidence: 99%

A New Method to Detect Buffalo Mastitis Using Udder Ultrasonography Based on Deep Learning Network

Zhang,

Li,

Zhang

et al. 2024

Animals

Self Cite

View full text Add to dashboard Cite

show abstract

“…In our multivariable model (Table 4), quarter SCS was significantly associated with NPV, PSD, Gradient Mean Value, Homogeneity, Gradient Variance and Entropy, whilst udder quarter and time were significant, too. The authors of [12] studied buffaloes and also found that echotexture features (NPV and PSD) were significantly affected by SCC, lactation stage, scanning plane and udder quarter. These results indicate that echotexture analysis could aid the detection of subclinical mastitis-induced alterations of the mammary parenchyma.…”

Section: Discussionmentioning

confidence: 99%

“…Echotexture analysis employs statistical features, such as mean Numerical Pixel Values (NPV) and Pixel Standard Deviation (PSD)-which are first-order echotexture features-to quantify parenchymal changes depicted in grey-scale images (B-mode sonograms) due to the presence of milk or under pathological conditions [10,11]. In buffaloes, first-order echotexture features were significantly correlated with milk somatic cell count (SCC) [12]. Furthermore, second-and higher-order echotexture features have been used in udder echotexture analysis [13].…”

Section: Introductionmentioning

confidence: 99%

Udder Ultrasonography of Dairy Cows: Investigating the Relationship between Echotexture, Blood Flow, Somatic Cell Count and Milk Yield during Dry Period and Lactation

Themistokleous

Papadopoulos

Panousis

et al. 2023

Animals

View full text Add to dashboard Cite

Udder health of dairy cows is related to their productivity and welfare. The period from dry-off to calving and early lactation is crucial. Ultrasonography is a useful and practical tool for the examination of the mammary parenchyma and blood flow. This observational study investigated the relationship between udder echotexture features, blood flow volume (BFVol) in the milk vein, milk somatic cell count (SCC) and daily milk yield (DMY) from late lactation, throughout the dry period and consecutive early lactation. Seventeen repeated measurements were performed on twenty-one Holstein cows. The udder parenchyma was examined with B-mode ultrasonography. Udder echotexture was studied using 15 features: Numerical Pixel Value (NPV), Pixel Standard Deviation (PSD), Skewness, Excess, Contrast, Homogeneity, Correlation, Entropy, Run Percentage, Long-Run Emphasis, Grey Value Distribution, Runlength Distribution, Gradient Mean Value, Gradient Variance and Percentage of Non-zero Gradients. Blood flow in the milk vein was examined with spectral Doppler. Linear mixed-effects models were employed to investigate relationships between BFVol, udder echotexture features, SCC and DMY throughout the study period. Our models showed that a 1 kg increase in DMY was associated with a significant increase of 0.25 L/min in the expected BFVol and that a 1,000,000-cells/mL increase in SCC was associated with a significant BFVol decrease of 0.49 L/min, keeping all other variables constant. Multivariable models showed significant associations between DMY and NPV, between PSD and Long-Run Emphasis, and between SCC and NPV, PSD, Gradient Mean Value, Homogeneity, Gradient Variance and Entropy. In conclusion, udder echotexture and BFVol in the milk vein are related to SCC and milk yield. Ultrasonography can be used for the comprehensive assessment of udder health in support of precision dairy farming.

show abstract

“…The image resolution as well as the sound waves' depth of penetration into the tissues depend on the probe's frequency (Williams, 2002;Silva et al, 2006;Fasulkov 2012); lower frequencies produce less resolution but have greater depth of penetration (Lucas et al, 2014). Various studies in large and small ruminants also evaluated the potential influence of the scanning plane on the echotextural characteristics of the mammary gland (Schwarz et al, 2020;Murawski et al, 2019;Zhang et al, 2022). However, there has been no information on the effect that scanning plane may have on the quantitative relationships between mammary echotexture and milk parameters in ewes.…”

mentioning

confidence: 99%

A Preliminary Study of the Relationships Between Echotextural Characteristics of the Mammary Gland and Chemical Composition of Milk During Early Lactation in EWES

Molik

JavadiEsfahani

Murawski

et al. 2023

Annals of Animal Science

View full text Add to dashboard Cite

This study aimed to examine the ultrasonographic characteristics of the mammary gland in post-partum ewes (1 to 4 weeks after lambing) of varying genotypes for correlations with the chemical composition of milk. Ultrasonograms of mammary glands in coronal (C) and sagittal (S) planes were obtained using 5.0- and 7.5-MHz transducers from four Polish Mountain (PMt) and six Olkuska (Olk) ewes. Computerized image analyses of all images have been performed to determine pixel intensity (numerical pixel values-NPV) and heterogeneity (pixel standard deviation-PSD) of the mammary gland stroma/parenchyma. In PMt ewes examined with the 5.0-MHz probe, the mean NPV-C correlated negatively with C17:0 (margaric acid) and C17:1 c10 (heptadecenoic acid). Furthermore, NPV-S correlated negatively with C17:0 content of milk. In PMt ewes examined with the 7.5-MHz transducer, the mean NPV-S correlated positively with the crude fat and total solids, and negatively with C18:3 c9, c12, c15 (n-3) (α-linoleic acid) content of milk samples, while PSD-S correlated positively with C16:0 (palmitic acid) and negatively with lactose. In the scans obtained with the 5.0-MHz in Olk sheep, the mean NPV-C was directly related to C10:1 (decanoic acid), C14:1 (myristoleic acid), C16:1 iso (iso-hexadecanoic acid), C16:1 c9 (palmitoleic acid), and C18:0 (stearic acid) content; the mean PSD-C correlated positively with total solids, C14:1, C15:0, C16:0, C16:1 iso, C16:1 c9, CLA (conjugated linoleic acids), and total saturated fatty acid content, and negatively with C18:1 c11 (n-7) (cis-vaccenic acid), C18:2 c9,c12 (n-6) (linoleic acid), and total polyunsaturated fatty acid content of milk; and the mean PSD-S correlated directly with ash and total saturated fatty acid content of the milk. Lastly, the mean NPV-C of images recorded with the 7.5-MHz probe in Olk ewes related directly to ash and C12:0 (lauric acid); the mean NPV-S related directly to ash and C16:1 iso; the mean PSD-C correlated positively with ash and total saturated fatty acid (SFA) content, and the mean PSD-S correlated positively with the total SFA. Echotextural attributes of the mammary gland in ewes have the makings of the technique for determining milk content of certain chemical constituents in a breed-specific manner; however, further studies are needed to verify the usefulness of echotextural analyses of the udder to consistently predict complete milk composition.

show abstract

Relationship Between Somatic Cell Counts and Mammary Gland Parenchyma Ultrasonography in Buffaloes

Cited by 4 publications

References 40 publications

A New Method to Detect Buffalo Mastitis Using Udder Ultrasonography Based on Deep Learning Network

A New Method to Detect Buffalo Mastitis Using Udder Ultrasonography Based on Deep Learning Network

Udder Ultrasonography of Dairy Cows: Investigating the Relationship between Echotexture, Blood Flow, Somatic Cell Count and Milk Yield during Dry Period and Lactation

A Preliminary Study of the Relationships Between Echotextural Characteristics of the Mammary Gland and Chemical Composition of Milk During Early Lactation in EWES

Contact Info

Product

Resources

About