Predicting pork freshness using multi-index statistical information fusion method based on near infrared spectroscopy

Area of Science:

• Food Science
• Analytical Chemistry
• Spectroscopy

Background:

• Pork freshness is crucial for food safety and quality.
• Traditional methods for assessing pork freshness can be time-consuming and destructive.
• Developing rapid, non-destructive techniques is essential for the meat industry.

Purpose of the Study:

• To apply near-infrared spectroscopy (NIR) for non-destructive pork freshness evaluation.
• To develop and validate a multi-index statistical information fusion (MSIF) modeling method for predicting pork freshness indicators.
• To compare the proposed MSIF method with existing variable selection techniques.

Main Methods:

• Utilized near-infrared spectroscopy (NIR) to collect spectral data from pork samples.
• Measured total volatile basic-nitrogen (TVB-N) and pH value as indicators of pork freshness.
• Developed a multi-index statistical information fusion (MSIF) model incorporating variable selection.
• Compared the performance of the MSIF model against other variable selection methods.

Main Results:

• The MSIF model demonstrated superior generalization performance and stability.
• Achieved a prediction correlation coefficient (Rval) of 0.8618 and RMSEP of 3.910 for TVB-N content.
• Achieved a prediction correlation coefficient (Rval) of 0.9379 and RMSEP of 0.1046 for pH value.
• The proposed method outperformed other state-of-the-art variable selection methods.

Conclusions:

• NIR technology combined with the MSIF model shows significant potential for rapid and non-destructive determination of pork freshness.
• This approach offers a promising tool for real-time meat quality monitoring.
• The study highlights the value of information fusion for enhancing analytical insights in the food industry.