A noninvasive near infrared system for detection of platelet components contaminated with bacteria.

BACKGROUND: Platelet (PLT) transfusion-associated bacterial sepsis has remained a substantial patient risk, primarily due to lacking effective and point-of-issue measures to detect bacterial contamination. This study describes near infrared (NIR) spectroscopy to examine inoculated PLTs without sampling within a few seconds. STUDY DESIGN AND METHODS: This study evaluated apheresis PLTs inoculated with low concentrations of Bacillus cereus and Staphylococcus epidermidis, comparing with sterile bags. Short-wavelength NIR spectra over the range from 700 to 1100 nm in the transmittance mode were obtained using research (NIRS6500, Foss NIRSystems) and portable (NIRGun, Shizuoka Shibuya Seiki) instruments at 6-hour intervals from 0 to 72 hours after inoculation (HAI). RESULTS: The sensitivity of the NIRS6500 was 100% (43/43) and 98% (50/51) after incubating PLTs inoculated with B. cereus for 42 HAI or more and with S. epidermidis for 54 HAI or more, respectively. Difference spectra calculated by subtracting the NIR spectra of stored PLTs with that of the same PLTs measured at 0 HAI improved the discrimination results compared with conventional second derivative spectra. CONCLUSION: The NIRS6500 system can provide a PLT monitoring system based on difference spectra. The chemical components of PLTs that were influenced by bacterial metabolism seemed to play an important role in the calibration structure. Further studies should examine samples spiked with various species of prevalent bacteria.

[Prediction of water holding capacity of fresh pork using near infrared spectroscopy].

Water holding capacity (WHC) is an important quality characteristic of fresh pork. It is necessary to determinate it by a rapid and nondestructive method in order to reduce the loss of meat. Near infrared spectroscopy as a new method was proposed for rapid and nondestructive measurement of WHC of vacuum-packed pork loin. Two reference methods for measuring water holding capacity were used to evaluate the water holding capacity, i. e. drip loss and press method. The acquired raw spectra were pretreated by Savisky-Golay smoothing and second derivative, respectively. A total of 106 samples were used in the experiment. The samples were divided into calibration set and validation set. The calibration set was used to set up calibration model and then the model was adopted to predict the samples of validation set. The partial least square regression (PLSR) was used to build calibration model. The results show that the correlation coefficient for drip loss and press method is 0.74-0.79. It shows that evaluating the water holding capacity of vacuum-packed fresh pork loin using near infrared spectroscopy in diffuse reflection mode is better than those results with reflectance or transmission.