Submitted to: Journal of Animal Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: July 26, 2011
Publication Date: December 21, 2011
Citation: King, D.A., Shackelford, S.D., Wheeler, T.L. 2011. Use of visible and near-infrared spectroscopy to predict pork longissimus lean color stability. Journal of Animal Science. 89:4195-4206. Interpretive Summary: Losses associated with discounting or discarding retail packages of pork loin chops due to discoloration are a significant burden to the industry. This is particularly true in large, “family sized” packages of pork loin chops where one discolored chop reduces the value of several acceptably colored chops. The present experiment was conducted to develop technology to sort pork loins at fabrication, after aging, and after chop cutting based on predicted retail color-life. Visible and near-infrared reflectance spectra were collected on pork loins immediately after back rib removal, after aging for 14 d, or directly on chops after cutting. Chops were then placed in simulated retail display for 14 d. All three prediction models effectively sorted loins into color stability groups. In general, loin chops predicted to have stable lean color had color attributes on d 14 of display that were similar to those of loins predicted to have labile lean color on d 7 of display. Thus, visible and near-infrared spectroscopy is a promising technology for sorting pork loins with regard to lean color stability.
Technical Abstract: This study evaluated using visible and near-infrared (VIS/NIR) spectroscopy to predict lean color stability in pork loin chops. Spectra were collected immediately following and approximately 1 h after rib removal from 1,208 loins. Loins were aged for 14 d before a 2.54-cm chop was placed in simulated retail display. Spectra were collected on aged loins immediately following removal from the vacuum package and on chops 10 min after cutting. Instrumental color measurements (L*, a*, b*, hue angle, chroma, and delta E) were determined on d 0, 1, 7, 11, and 14 of display. Principal component analysis of display day 0 and 14 values of these traits identified a factor (PC1) explaining 67% of the variance which was related to color change. Partial least squares regression was used to develop three models to predict PC1 values using VIS/NIR spectra collected in the plant, on aged loins, and on chops. Loins with predicted PC1 values less than 0 were classified as having stable color, while values greater than 0 were classified as having labile lean color. Loins classified as stable by the in-plant model had lower (P<0.05) L* values than those classified as labile. Hue angle and delta E were lower (P<0.05) and a* and chroma values were greater (P< 0.05) after d 7 of display in loins predicted to have stable color than loins predicted to have labile lean color. Similarly, chops from loins classified as stable using the aged loin model had lower (P<0.05) L* values than those from loins classified as labile. Furthermore, loins predicted to be stable had lower (P < 0.05) hue angle and delta E values and higher (P<0.05) a* and chroma values after d 7 of display than loins predicted to be labile. Results for the chop model were similar to those from the two loin models. Chops predicted to have stable lean color had lower (P<0.05) L* values than those predicted to have labile lean color. Chops classified as stable had lower (P<0.05) hue angle and delta E values and higher (P<0.05) a* and chroma values after d 7 of display than chops classified as labile. All three models effectively segregated chops based on color stability, particularly with regard to redness. Regardless of the model being used, d 14 display values for a*, hue angle, and delta E in loins classified as stable were similar to the d 7 values of loins classified as labile. Thus, these results suggest that VIS/NIR spectroscopy would be an effective technology for sorting pork loins with regard to lean color stability.