Detection of woody breast condition in commercial broiler carcasses using image analysis

Image analysis could be an objective and rapid method to identify woody breast (WB) myopathy and benefit the global poultry industry. The objective of this study was to determine if there are conformational changes that can be used to detect WB characteristics in commercial broiler carcasses across strains, gender, and ages using image analysis. A total of 900 images of male and female broiler carcasses from commercial standard and high breast-yielding strains and 5 ages (6 through 10 wk) were captured before evisceration. These images were processed and analyzed using ImageJ software. Conformational measurements were M0: breast length; M1: breast width in the cranial region; M2: vertical line from the tip of keel to 1/5th of breast length; M3: breast width at the end of M2; M4: angle formed at the tip of keel and extending to outer points of M3; M5: area of the triangle formed by M3 and lines generated by M4; M6: area of the breast above M3; M7: M6minus M5. Ratios of these measurements were also considered. Intact breast fillets were scored for WB severity based on tactile evaluation. Regardless of strain, sex, and age, M11 (M1/M0), M9 (M3/M2), and M4 had the highest correlation to WB score (r(s) >= 0.65; P < 0.01). Overall, the best validated model (Gen. R-2 = 0.61) to predict WB included M1, M2, and M3. Using this model, 91% of broiler carcasses were properly classified as normal or WB along with a sensitivity of 71% to detect affected carcasses. Although the predictive performance of models for detecting the WB condition using these measurements was associated with the broiler strain, sex, and age or live weight, these data also support the feasibility of using image analysis to predict WB defect in broiler carcasses. The possible integration of these image measurements into commercial noncontact, nondestructive, and fast in-line vision grading systems would allow processors to identify broilers with WB and potentially sort, provide large-scale information downstream to further processing operations and upstream to live production.