Project : USDA ARS

ARS Home » Southeast Area » Athens, Georgia » U.S. National Poultry Research Center » Quality and Safety Assessment Research Unit » Research » Research Project #439244

Research Project: Assessment of Quality Attributes of Poultry Products, Grain, Seed, Nuts, and Feed

Location: Quality and Safety Assessment Research Unit

2024 Annual Report

Objectives
1. Assess the intrinsic properties of myopathic chicken that alter quality and processing attributes in meat and enable commercially-viable processing strategies to limit myopathy impact. 1.A. Identify the mechanisms by which the physical and chemical properties of myopathic broiler muscles influence quality and processing attributes of meat products. 1.B. Evaluate processing and formulation strategies to minimize the negative impact of broiler muscle myopathies on the technological, compositional, and sensory properties of meat products. 2. Develop nondestructive, rapid imaging technologies to enable commercial measurement of quality characteristics and defects in poultry meat and eggs. 2.A. Develop a high-speed imaging technology for detecting and sorting poultry muscle myopathies and meat quality defects. 2.B. Utilize sensor fusion to enhance the ability of imaging technology to simultaneously assess multiple quality attributes and defects in poultry meat. 2.C. Develop imaging technology for rapid assessment of egg quality and defects. 3. Develop rapid, nondestructive microwave sensors to enable commercial measurement of quality parameters in grain, seed, nuts, and feed. 3.A. Enable distributed networks of microwave sensors for real-time monitoring of moisture content in grain, seed, and nut storage facilities. 3.B. Enable on-the-trailer multiparameter microwave sensors for nondestructive and instantaneous grading and monitoring of drying nuts. 3.C. Enable microwave sensors for simultaneous and nondestructive determination of moisture content and water activity of peanuts, almonds, and other nuts.

Approach
Poultry meat, egg, grain, seed, nut and feed commodity values depend upon quality. Research on poultry meat quality defects will focus on underlying mechanisms, utilization methods and rapid detection/sorting systems for quality defects. To determine how woody breast (WB) affects postmortem changes in breast muscle, trials will measure WB muscle deboning response, rigor mortis development and postmortem energy metabolism. Low field time-domain nuclear magnetic resonance techniques will be used to assess how muscle water properties influence WB meat quality during aging, cooking, freezing and marination. Impact of spaghetti meat (SM) on breast meat quality, composition, and functionality will be measured. Effects of SM on processing and quality in further-processed products (ground meat patties, fresh sausages and hotdogs) will be measured. A machine vision WB detection technology will be expanded to integrate the side-view imaging component into a system for both detection and sorting. Image acquisition and processing will be enhanced to match commercial processing line speeds and system will be tested on breast meat from a range of broiler varieties and sizes. To simultaneously assess multiple quality traits associated with WB and white striping, sensor fusion techniques will be evaluated. Multiple sensors measuring 2D and 3D shape morphology, spatial texture, muscle rigidity, color, and spectral data will be evaluated via independent trials. Once best sensing modalities are determined, sensor fusion algorithms will be developed and tested. For measuring egg quality, a modified-pressure imaging system to detect hairline cracks will be modified to grade table eggs for air-cell depth and yolk shadow using new machine vision algorithms. System will be redesigned for online operation and applied to detect cracked eggs in hatcheries. Microwave sensors for quality assessment of grains, seeds, nuts, and feeds will be developed. Microwave sensors in a distributed network will be developed for real-time nondestructive monitoring of moisture content in storage facilities for peanuts, almonds, wheat, corn, soybean and corn-soybean meal. Following laboratory testing in an eighth-scale drying bin equipped with multiple sensors, sensor networks will be tested in commercial grain and nut facilities. Microwave sensors will be developed to assess multiple attributes (moisture, bulk density, meat content and foreign material) before and during drying of peanuts, almonds, pecans and pistachios. After calibration with static samples, sensors will be tested in a quarter-scale nut drying system. Microwave sensors will be developed to simultaneously measure moisture content and water activity of in-shell peanuts, almonds and other nuts. A dielectric database will be collected with lab grade instrumentation. Following selection of optimal frequencies, prototype sensors will be assembled, calibrated and tested. By seeking to understand quality attributes, investigating utilization methods and developing rapid assessment tools, this project takes a multifaceted approach to provide information and technologies for producing and marketing high quality commodities.

Progress Report
Multiple studies were conducted to identify the mechanisms by which the physical and chemical properties of myopathic broiler muscles influence quality and processing attributes of meat products (Sub-objective 1A). Studies were conducted to understand: 1) how myopathies develop in the muscle, and 2) how myopathies influence meat quality attributes during postmortem processing and handling. Research on targeted verification of differentially abundant proteins, metabolites, and lipids identified in spaghetti meat (SM) breast fillets compared to normal fillets was conducted using biochemical assay systems. Differential abundances for several factors were confirmed. Potential genetic variations were identified using massively parallel DNA sequencing methods. Trials to characterize the effects of the wooden breast (WB) myopathy on mitochondria function in various tissues and metabolic pathways in muscle were conducted. On the postmortem side, initial trials were conducted to determine the effects of WB on early postmortem muscle metabolism. For these trials, muscle samples were collected at various times postmortem. Biochemical analysis for these trials is ongoing. A study was completed to determine the metabolomic differences between normal breast meat, SM, and WB meat at 24 h postmortem. Trials were also conducted on the effects of postmortem rigor mortis development and sarcomere length (i.e., muscle shortening) on broiler breast meat with the WB condition. An experiment was completed using low-field nuclear magnetic resonance (LF-NMR) techniques to delineate changes in myowater distribution in WB meat due to freezing-thawing (three freezing temperature, -20, -40, and -200°C and two thawing temperature, 4 and 20°C) and their influence on quality attributes. Preliminary data was collected on the histological differences observed in normal and WB tissue due to freezing/thawing. Research was conducted to evaluate processing and formulation strategies to minimize the negative impact of broiler muscle myopathies on the technological, compositional, and sensory properties of meat products (Sub-objective 1B). Research trials were completed to determine the sensory and compositional effects of including an increasing proportion of SM into the formulation of a chicken-nugget type product. Additionally, as part of this study myowater properties of the products were measured using LF-NMR as a preliminary investigation into the potential for using LF-NMR techniques as a non-destructive, rapid test to determine the sensory/organoleptic attributes of meat products. Data analysis was completed for multiple experiments designed to determine the technological and sensory effects of including myopathic breast meat (SM and WB) into the formulations of frankfurters. Preliminary trials were also conducted to compare the antioxidant capacity and contents of functional small peptides and amino acids in hydrolysates generated from normal and WB meat. As part of collaborative projects with several universities, several projects were completed to determine the effects of broiler processing parameters on meat quality. Trials were completed in a commercial broiler processing plant to determine the meat quality effects of delayed broiler carcass processing. Research was also conducted to determine the effects of broiler stunning method and deboning time on muscle proteins and breast meat quality. Research was conducted to develop a high-speed imaging technology for detecting and sorting poultry muscle myopathies and meat quality defects was conducted (Sub-objective 2A). Research built upon an imaging system previously developed at ARS that uses a side-view camera to rapidly and accurately detect WB meat based on how the breast fillets bend while traveling off a conveyor belt. Using the current configuration, it was determined that the system could only check one breast fillet at a time and would potentially require the installation of an additional camera on the opposite side of the conveyor belt to keep up with the high throughput on commercial poultry processing lines. An alternative approach was explored to increase throughput, a 3D machine vision technique using a single camera positioned to look down on the conveyor belt. Initial research progress demonstrated the potential of using a top-view 3D imaging approach to detect the WB condition in multiple fillets at once with higher accuracy. Research is ongoing to further develop this approach as it will potentially lead to a simpler and more accurate imaging system design, compared to the current side-view vision technology. Research was conducted to utilize sensor fusion to enhance the ability of imaging technology to simultaneously assess multiple quality attributes and defects in poultry meat (Sub-objective 2B). Research was advanced on testing new sensor technologies for better analysis of the WB condition in poultry breast meat. Trials were conducted on a sensor fusion method combining optical coherence tomography (OCT) and hyperspectral imaging. The OCT method was used to create detailed maps of the internal muscle structure, reaching depths of about 1 mm in broiler breast muscle tissue while hyperspectral imaging was used to map the surface fat content and meat composition of the breast meat. Findings indicated that a non-destructive OCT imaging modality has the potential to provide rapid, objective, and accurate assessment of the WB condition in boneless, skinless broiler breast fillets. Research also advanced on the development of a wearable tactile sensing glove for WB detection measurements. After testing different force sensor types and placement, a glove prototype equipped with multiple force sensors was developed and tested to evaluate how human experts subjectively assessed the WB condition by touching and handling the chicken breast fillets. Initial analyses were conducted on mimicking the human palpation scoring process and to extract valuable scientific information that can be used to design a robotic manipulator capable of "digital palpation" for automated WB detection. Research to develop rapid, nondestructive microwave sensors to enable commercial measurement of quality parameters in grain, seed, nuts, and feed was conducted (Objective 3). A compact system with two x-band focused-beam antennas connected to a vector network analyzer was calibrated and tested. This system allows for the computation of dielectric properties from reflection and/or transmission coefficients, thus providing application flexibility. The system was tested on grain and seed samples of varying moisture contents and bulk densities. A low-cost commercial miniature vector network analyzer (miniVNA Tiny) operating at 1-3000 MHz was assessed for measurements of reflection and transmission coefficients of grain/seed samples placed between two inexpensive flat microstrip antennas. Research was conducted to enable distributed networks of microwave sensors for real-time monitoring of moisture content in grain, seed, and nut storage facilities (Sub-objective 3A). Wireless protocols (WLAN, Wi-Fi, ZigBee, and Bluetooth) were assessed and Wi-Fi was selected as the optimal mode of communication for the sensor network. Thingspeak, an Internet of Things (IoT) application, was used as the application programming interface for data to be viewed remotely in real-time from each sensor. Custom software was written to control each sensor using an embedded Raspberry Pi single board computer instead of an external PC connected by USB. Real-time measurements of dielectric properties within an eighth-scale grain drying bin were used to determine moisture content, density, shrinkage, mass reduction, and bed height of wheat, corn, and soybean during drying. Research was conducted to enable on-the-trailer multiparameter microwave sensors for nondestructive and instantaneous grading and monitoring of drying nuts (Sub-objective 3B). The original microcontroller used for the sensors was replaced with a Raspberry Pi single-board computer to allow remote connection and real-time data retrieval. Custom software was written to facilitate dielectric measurements. Sensors were equipped to measure temperature and relative humidity within the air plenum and peanut bed and of the exhaust air on trailers. Reference methods for oven moisture content determination for certain in-shell nuts were established. Verification studies were carried out to assess calibration stability and accuracy of the sensors to determine in-shell kernel moisture content. Microwave sensors were compared in parallel with the convection oven and the official meter. Multiple studies were conducted to enable microwave sensors for simultaneous and nondestructive determination of moisture content and water activity of peanuts, almonds, and other nuts (Sub-objective 3C). A method was developed and tested for determining multiple quality attributes of in-shell peanuts and almonds including bulk density, moisture content and water activity from the measurement of dielectric properties at a single microwave frequency. Dielectric properties of in-shell and shelled walnuts of varying bulk density and moisture content were measured with a free-space transmission technique in the frequency range 2-18 GHz and temperature 0-45oC. Commercial versions of the in-shell kernel moisture prototype sensor were tested at three peanut buying points in Georgia. Over 2,000 data points were collected. As part of collaborative work with another ARS lab (Stillwater, OK), peanut pods were modified with different fillings and AI models were strengthened to improve classification accuracy based on measurements of their dielectric properties.

Accomplishments
1. Further-processed meat products made from myopathic chicken breast meat. Due to problems with muscle integrity and poor visual appeal, broiler breast meat exhibiting the quality defect known as spaghetti meat (SM) myopathy is not suitable for whole muscle products. ARS researchers and university collaborators in Athens, Georgia, conducted a series of studies to demonstrate that chicken breast meat with the SM myopathy can be included in the formulations of products such as ground meat patties, frankfurters, and chicken nuggets at low to moderate levels without negatively influencing processing yields or sensory attributes. These findings indicate that the broiler industry can utilize SM in further-processed meat products to recoup lost value without compromising product quality.

2. Machine learning and AI-based software for moisture determination in grain and seed. Microwave sensing has many potential applications for rapidly and nondestructively measuring moisture content in a variety of agricultural commodities. However, the development of microwave-based sensors requires empirically measuring the effects of many input variables on dielectric properties and extensive spectra data modeling. ARS researchers in Athens, Georgia, developed a machine learning and artificial intelligence (AI)-based software to accurately predict grain and seed moisture content from their dielectric properties measured at a single microwave frequency. This software has the advantage of providing moisture content without making any assumptions concerning the electromagnetic wave used for sensing and material interaction or the need for previously established analytical models correlating the dielectric properties to moisture content. Another important feature of the software is that moisture content can be determined with or without knowledge of bulk density of the material being measured. This software provides a critical tool for developing microwave moisture sensors for a variety of applications.

3. Muscle water properties influence meat quality traits in myopathic chicken breast meat. The wooden breast (WB) myopathy is a meat quality defect that occurs in the breast muscle of fast-growing broiler chickens and is known to result in breast meat with abnormal texture and inferior water-binding ability. How tightly or loosely water is bound or entrapped within the microstructure of muscle cells is thought to influence meat quality traits. Through a series of experiments, ARS researchers in Athens, Georgia, used low-field nuclear magnetic resonance (LF-NMR) methods to demonstrate that changes in muscle water properties change during the first 24 h postmortem and are different between normal and WB meat. Findings suggest for the first time that high levels of free water within the muscle tissue are closely related to the poor water-holding capacity and unique texture properties of WB meat. Understanding these underlying mechanisms is necessary to develop more effective postmortem processing and handling techniques to minimize meat quality problems due to the presence of WB meat.

4. Ribonucleotide reductase enzyme in wooden breast (WB) chicken. The underlying biological mechanisms that cause the wooden breast (WB) myopathy and associated meat quality defects in broiler breast muscle are not well understood. ARS researchers in Athens, Georgia, demonstrated for the first time that ribonucleotide reductase (an enzyme necessary for DNA synthesis and repair) and genes related to mitochondria function are altered in WB meat. These findings suggest that energy metabolism in affected muscle tissue is impaired and provides a key insight on potential pathways leading to WB development.

U.S. DEPARTMENT OF AGRICULTURE

Quality and Safety Assessment Research Unit: Athens, GA