Performance of human observers and an automatic 3-dimensional computer-vision-based locomotion scoring method to...

The objective of this study was to determine if a 3-dimensional computer vision automatic locomotion scoring (3D-ALS) method was able to outperform human observers for classifying cows as lame or nonlame and for detecting cows affected and nonaffected by specific type(s) of hoof lesion. Data collection was carried out in 2 experimental sessions (5 mo apart). In every session all cows were assessed for (1) locomotion by 2 observers (Obs1 and Obs2) and by a 3D-ALS; and (2) identification of different types of hoof lesions during hoof trimming (i.e., skin and horn lesions and combinations of skin/horn lesions and skin/hyperplasia). Performances of observers and 3D-ALS for classifying cows as lame or nonlame and for detecting cows affected or nonaffected by types of lesion were estimated using the percentage of agreement (PA), kappa coefficient (κ), sensitivity (SEN), and specificity (SPE). Observers and 3D-ALS showed similar SEN_lame values for classifying lame cows as lame (SEN_lame comparison Obs1-Obs2 = 74.2%; comparison observers-3D-ALS = 73.9-71.8%). Specificity values for classifying nonlame cows as nonlame were lower for 3D-ALS when compared with observers (SPE_nonlame comparison Obs1-Obs2 = 88.5%; comparison observers-3D-ALS = 65.3-67.8%). Accordingly, overall performance of 3D-ALS for classifying cows as lame and nonlame was lower than observers (Obs1-Obs2 comparison PA_lame/nonlame = 84.2% and κ_lame/nonlame = 0.63; observers-3D-ALS comparisons PA_lame/nonlame = 67.7-69.2% and κ_lame/nonlame = 0.33-0.36). Similarly, observers and 3D-ALS had comparable and moderate SEN_lesion values for detecting horn (SEN_lesion Obs1 = 68.6%; Obs2 = 71.4%; 3D-ALS = 75.0%) and combinations of skin/horn lesions (SEN_lesion Obs1 = 51.1%; Obs2 = 64.5%; 3D-ALS = 53.3%). The SPE_nonlesion values for detecting cows without lesions when classified as nonlame were lower for 3D-ALS than for observers (SPE_nonlesion Obs1 = 83.9%; Obs2 = 80.2%; 3D-ALS = 60.2%). This was translated into a poor overall performance of 3D-ALS for detecting cows affected and nonaffected by horn lesions (PA_{lesion/nonlesion} Obs1 = 80.6%; Obs2 = 78.3%; 3D-ALS = 63.5% and κ_{lesion/nonlesion} Obs1 = 0.48; Obs2 = 0.44; 3D-ALS = 0.25) and skin/horn lesions (PA_{lesion/nonlesion} Obs1 = 75.1%; Obs2 = 75.9%; 3D-ALS = 58.6% and κ_{lesion/nonlesion} Obs1 = 0.35; Obs2 = 0.42; 3D-ALS = 0.10), when compared with observers. Performance of observers and 3D-ALS for detecting skin lesions was poor (SEN_lesion for Obs1, Obs2, and 3D-ALS <40%). Comparable SEN_lame and SEN_lesion values for observers and 3D-ALS are explained by an overestimation of lameness by 3D-ALS when compared with observers. Thus, comparable SEN_lame and SEN_lesion were reached at the expense high number of false positives and low SPE_nonlame and SPE_nonlesion. Considering that observers and 3D-ALS showed similar performance for classifying cows as lame and for detecting horn and combinations of skin/horn lesions, the 3D-ALS could be a useful tool for supporting dairy farmers in their hoof health management.