Influence of Thermal Treatments on Textural and Rheological Properties of Different Types of Meatballs

Featured Application

Textural and rheological investigations of meatballs derived from three different varieties of tenderloin enriched with herbs extracts processed through two milder heat treatments.

Abstract

Ready-to-eat products are very popular and controversial due to their microbial safety. The main processing steps in obtaining a safe, edible product is heat treatment. The traditional manufacturing of meatballs, which conducts unhealthy compounds related to deep-fat-fried foods like the oil oxidation of harmful substances and polycyclic aromatic hydrocarbons, has been replaced with baking (180 °C) and steaming (94 °C). The addition of aqueous extract from two herbs, lemon balm (Melissa officinalis L.) or wild thyme (Thymus serpyllum L.), has led to twelve variants of meatballs, obtained from the tenderloin of three different animal species (pork, turkey, and beef). During processing, the food components go through conformational changes that affect the texture of the final product. In this study, differential scanning calorimetry for detecting and characterizing the thermal changes in meatballs was used. In addition, the influence of heat treatments on the textural and rheological parameters of meatballs was evaluated using instrumental methods. The cooking yield registered values of 61.21 ± 0.25% for steamed beef samples and 81.36 ± 0.86% for steamed turkey samples. The latest samples also showed the lowest firmness value, 3.41 ± 0.79 N. In this study, the addition of aqueous extracts did not considerably affect the texture and rheological behavior, which were influenced mainly by the heat treatment and meat type. Generally, steaming determined a firmer texture compared to baking.

1. Introduction

Meat and meat-based products should be cooked before consumption to destroy foodborne pathogens and to increase their palatability. This represents a way to ensure microbial safety and to obtain a more edible and digestible product [1]. For example, to ensure the safe consumption of chicken meat, it should be heated at least to an internal temperature of 72 °C, but this leads to changes in the microstructure and texture of the meat. Therefore, due to all the changes in the microstructure, texture, and appearance of meat, the consumer’s acceptance might be affected [2]. Most of the thermal treatments used in the meat industry rely on conductive, convective, and radiative mechanisms from the heating medium (air, water, or oil) to the meat product. Depending on the characteristics of the product (the type, dimension, and geometry), it may take considerable time to conduct sufficient heat into the product core to reach the optimum temperature; therefore, some parts of the product might be overcooked or undercooked, which will adversely affect the product quality [3]. The meat texture is one of the most important quality attributes for the consumer and it is mainly influenced by protein denaturation, which leads to fiber shrinkage and straightening [4].

Generally, meatballs are minced meat derivatives containing salt, seasonings, and backfat, and other adding materials may be considered. In general, the raw meat is chopped and mixed first, and the ingredients are added after, which could maximize the extraction of salt-soluble proteins, with direct influence on the textural and rheological properties of the final products [5]. The color and aroma characteristics are determined by the chosen type of thermal treatment due to the specific properties of each thermal treatment.

Frying constitutes one of the most used thermal treatments, especially when talking about the traditional Romanian meatballs. But, considering the novel nutritional healthy trends, predominantly addressed to special destination foods, a significant concern has focused on reducing the fat content, in particular, in processed meats.

Lipid oxidation, which happens during the production and storage of meat or meat products, is another significant issue. The color, flavor, texture, and nutritional content of the products are all negatively impacted by the modifications this process generates. Phenolic antioxidants seem to be the most efficient inhibitors when it comes to stopping lipid oxidation. By default, plants, a natural source of antioxidants, as an ingredient in meatballs, could contribute to the reduction in the chemical ingredients commonly used in the processes sustaining the strategies in the production of healthy food [6].

Herbs and spices, leaves or extracts, have been added to food since ancient times to enhance its sensory qualities and increase its shelf life [7].

Still, the literature does not present data regarding the rheological and instrumental texture of meatballs, data that would be important for food engineers. It is expected that these characteristics will not be substantially affected by the aqueous extracts’ addition, but this consideration has to be validated.

The present study was designed to evaluate the texture development in pork, turkey, or beef tenderloin meatballs under the effect of two types of heat treatments (air convection and steam convection). Air convection and steaming are two healthy and common methods to prepare foods, especially by using two harmless and environmentally friendly processing agents represented by hot air and steam.

The meatball formulations were enriched with aqueous extracts of wild thyme and lemon balm to enhance their functionality and to validate that their addition will not affect the specific rheological and textural behavior of this kind of traditional dish.

The practical implications of this study are related to the development and characterization of texture, which is very important for this type of food product, influencing the acceptance of the consumers. Texture is a complex property, highly influenced by the type of meat, the formulation of the product, and the parameters of the heat treatment. It is also difficult to be described in an objective manner, since consumers have personal preferences. The results of this study present some possibilities to achieve meatballs and to characterize them impartially.

2. Materials and Methods

2.1. Organic Materials

Aerial parts of lemon balm (Melissa officinalis L.) and wild thyme (Thymus serpyllum L.) were purchased from an organic store in Galați, Romania. The plants were stored in a well-closed place, away from light and humidity until extraction. The meat was purchased from a supermarket in Galați, Romania, on each determination day and kept under refrigeration conditions (4 °C) until further processing.

2.2. Sample Preparation

a.

Preparation of the aqueous extracts of herbs

Aqueous extracts of lemon balm and wild thyme were obtained after the slightly modified method presented by Hmidani et al. [8] and the process is presented earlier in [9]. The extracts were chosen based on the method of extraction yield and their phytochemical properties, especially antioxidant activity.

The aqueous extracts were stored at 4 °C until meatballs’ preparation. Briefly, 5 g of ground (Gorenje grinder SMK150B, Ljubljana, Slovenia) herbs was mixed with 125 mL of bidistilled water. The mixtures were boiled for 30 min using a water bath (Nahita, 6 L, Izmir, Turkey), then filtered and cooled at 4 °C.

b.

Preparation of meatballs

Meatballs were prepared using the same anatomical portion (tenderloin) from three different types of meat: pork, turkey, and beef. The meat was minced (food processor Philips HR 7766, Székesfehérvár, Hungary) and mixed with aqueous extract of lemon balm/wild thyme (6%), sunflower oil (6%), salt (0.5%), and pepper (0.3%). The meatballs were manually modeled into 20 ± 1 g meatballs (3.5 ± 0.1 cm diameter) and subjected to heat treatment. Their spherical shape was selected based on the shape of the Romanian traditional meatballs consumed in our country.

The sample encoding was determined by taking into consideration the type of meat, herbal extract, and cooking type, as follows in

Table 1. Meatballs formulations and codifications.

Samples Code	Meat Type	Aqueous Extract Type	Type of Thermal Treatment
ECPC	Minced pork tenderloin	wild thyme	Hot air convection (baking)
ERPC	Minced pork tenderloin	lemon balm	Hot air convection (baking)
ECPA	Minced pork tenderloin	wild thyme	Water vapor convection (steaming)
ERPA	Minced pork tenderloin	lemon balm	Water vapor convection (steaming)
ECCC	Minced turkey tenderloin	wild thyme	Hot air convection (baking)
ERCC	Minced turkey tenderloin	lemon balm	Hot air convection (baking)
ECCA	Minced turkey tenderloin	wild thyme	Water vapor convection (steaming)
ERCA	Minced turkey tenderloin	lemon balm	Water vapor convection (steaming)
ECVC	Minced beef tenderloin	wild thyme	Hot air convection (baking)
ERVC	Minced beef tenderloin	lemon balm	Hot air convection (baking)
ECVA	Minced beef tenderloin	wild thyme	Water vapor convection (steaming)
ERVA	Minced beef tenderloin	lemon balm	Water vapor convection (steaming)

.

2.3. Cooking Methods and Cooking Measurements

In the preliminary experiments of this study, the cooking parameters were determined by several treatments according to the current scientific literature. Therefore, several temperatures and time combinations were performed. The meatballs were subjected to two types of heat treatment: hot air convection and water vapor convection. Hot air convection was carried out at 180 °C using Indesit Electric Oven (FIMB-51K.A-IX-PL, Poland). The pork meatballs were treated for 36 min; meanwhile, the turkey and beef meatballs were treated for 30 min. In the case of water vapor convection, all the meatballs were treated for 15 min at 94 °C using a steam cooker (Zelmer 37Z010, Poland). The thermic treatment was applied until a temperature of at least 85 °C was reached in the thermal center of the product. For each sample, the mass, diameter, core temperature, and dry substance (Kern MRS 120-3, Germany) were measured before and after the heat treatment.

Cooking loss, cooking yield, moisture retention, and the reduction in meatball diameter were calculated according to [10], using the equations below:

$$\mathrm{C}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{s}\left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{U}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}-\mathrm{C}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}{\mathrm{U}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}}\times 100$$

(1)

Cooking yield was determined by measuring the weight of the samples before and after cooking.

$$\mathrm{C}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{C}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}{\mathrm{U}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}}\times 100$$

(2)

$$\mathrm{M}\mathrm{o}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e}\mathrm{r}\mathrm{e}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{\%}\mathrm{Y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\times \mathrm{\%}\mathrm{M}\mathrm{o}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e}\mathrm{i}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}}{100}}$$

(3)

$$\mathrm{R}\mathrm{e}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{d}\mathrm{i}\mathrm{a}\mathrm{m}\mathrm{e}\mathrm{t}\mathrm{e}\mathrm{r}\left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{U}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{d}\mathrm{i}\mathrm{a}\mathrm{m}\mathrm{e}\mathrm{t}\mathrm{e}\mathrm{r}-\mathrm{C}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{d}\mathrm{i}\mathrm{a}\mathrm{m}\mathrm{e}\mathrm{t}\mathrm{e}\mathrm{r}}{\mathrm{U}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{b}\mathrm{a}\mathrm{l}\mathrm{l}\mathrm{d}\mathrm{i}\mathrm{a}\mathrm{m}\mathrm{e}\mathrm{t}\mathrm{e}\mathrm{r}}}\times 100$$

(4)

whereas the reduction in diameter of the meatballs was measured using a digital caliper [mm].

2.4. Chemical Composition of Raw Meats and Meatballs

The chemical composition of each sample of raw meat (pork, turkey, and beef) and meatballs was analyzed according to the AOAC (2012) standard.

Fat content was determined by the Soxhlet method using a solvent extraction system (Soxhlet–Randell–Twisselmann, Raypal, Spain), while the protein content was determined with an automatic Kjeldahl nitrogen analyzer (Scitek Global Co., Jinan, China). Ash content was determined at a very high temperature (>600 °C) to remove all organic material using a muffle furnace (Nabertherm, Lilienthal, Germany).

The moisture content was determined by gravimetrically drying using a moisture analyzer (Kern MRS 120-3, Kern & Sohn GmbH, Germany).

All meatballs were subjected to textural, rheological, and calorimetric analysis and finally to sensory evaluation to determine if there were significant (p < 0.05) differences between the two thermal treatments used.

2.5. Texture Profile Analysis

The texture analysis of the samples was performed by applying the TPA (Texture Profile Analysis) method, which consists of a double penetration, using Brookfield CT3 texture analyzer (AMETEK Brookfield, Middleborough, MA, USA). The samples were divided into cylindrical pieces, with a diameter of 35 mm and a height of 12 mm. A metal cylinder with a diameter of 4 mm was used for the double penetration of the samples, up to a depth of 5 mm, with a speed of 1 mm/s. The textural parameters (firmness, adhesiveness, cohesiveness, springiness, chewiness, and gumminess) were determined with TexturePro CT V1.5 software, provided by Brookfield Engineering Labs. Inc., (Middleborough, MA, USA). For each sample, four determinations were performed and the results are presented as mean ± standard deviation (SD).

2.6. Rheological Analysis

The rheological characterization of the samples was carried out by performing creep–recovery tests using a stress-controlled rheometer (AR2000ex, TA Instruments, Ltd., New Castle, UK). The temperature was set at 25 °C using a Peltier temperature control system, with a plate–plate geometry of 40 mm diameter and a 2 mm gap.

The procedure was conducted using a parallel plate geometry with a 40 mm diameter and a gap of 2 mm. Meatballs were sliced with a special die allowing them to achieve the needed dimensions. For the creep step, the constant stress of 50 Pa was applied for 5 min. A stress-sweep test was preliminarily performed to ensure that the creep tests were carried out in the linear viscoelastic domain. For the recovery step, the stress was set at 0 Pa, allowing the sample to recover for 10 min.

2.7. Differential Scanning Calorimetry

The thermal measurements were performed by using Differential Scanning Calorimeter (DSC) for specific heat analysis; meanwhile, the acquisition and evaluation of data were determined with STAR^e software DSC3. The samples were subjected to five heating–cooling cycles in temperature interval of −20 °C to 150 °C and two isothermal processes with 2 min duration at each −20 °C and 150 °C temperatures. The cooling rate was 10 °C/min; meanwhile, the heating rate was 6 °C/min between −20 °C and 50 °C, 2/min between 50 °C and 80 °C, and 1 °C/min between 80 °C and 150 °C.

2.8. Sensory Evaluation

This research was approved by “Dunărea de Jos” University of Galati Ethics Committee (UEC ethics approval number 23/30.04.2025). The sensorial analysis was conducted in a standard sensory laboratory located in the Faculty of Food Science and Engineering, “Dunărea de Jos” University of Galati. An acceptance test using a 9-point hedonic scale (1 represents dislike extremely, while 9 represents like extremely) was carried out to analyze the sensory properties of meatball samples. The samples of the products were evaluated by a group of 10 panellists (8 women and 2 men, age 24–65). The panelists were chosen on the basis of availability, sensitivity, willingness, and good health. Three training sessions were given to introduce panelists to the products and their sensory characteristics previous to final sensory analysis. The sensory evaluation was carried out according to SR EN ISO 5492:2009 and SR EN ISO 8586:2014. All samples of cooked meatballs were blind coded with 3-digit random numbers, served in random order, and presented on a white dish to each panellist. The sensorial test was carried out in individual booths, at a controlled temperature of 21 ± 1 °C, using fluorescent lamps (simulating daylight). Plain water was offered to the panellists to rinse their mouth between tasting the meatballs samples. Each meatball sample was evaluated using the following descriptors: appearance, taste, aroma, aftertaste, mouthfeel, firmness, cohesiveness, juiciness, and overall acceptability. Principal component analysis (PCA) is a multivariate statistical method used to process and organize data distribution. This technique generates new data that are built based on covariance, eigenvalues, and eigenvectors from the selected data. The PCA method groups the data covariance by ordering eigenvalues from the highest to the lowest, in the order of significance. PCA was used to measure the human perception or behavior by reducing a great number of data points to a smaller number of data points using the principal components with an eigenvalue higher than 1 [11] and a cumulative variance greater than 60% [12]. Principal Component Analysis (PCA) was performed to study the differences and similarities between meatball samples.

2.9. Statistical Analysis

All analyses were conducted in at least three replicates and data are reported as mean ± SD. To identify significant differences, experimental data were subjected to one-way analysis of variance (ANOVA), using Minitab 19 statistical software (Romsym Data, Bucharest, Romania) followed by the Tukey test. Differences were considered as significant when p values were <0.05.

Before starting the statistical analysis, a check of distribution of numerical values was realized to detect if there were some anomalies that would lead to a loss in the significance of the obtained results. Thus, a chi-squared test (χ²) was performed.

Principal Component Analysis (PCA) was used to evaluate the relationship between sensory characteristics, textural parameters, and cooking measurements using the mean values. Hierarchical Cluster Analysis (HCA) was performed to identify similar meatball groups based on the obtained results. Partial Least Square (PLS) regression was also applied to model the relation between textural parameters (independent variables) and sensory characteristics (dependent variables) of meatballs with aqueous extract of lemon balm/wild thyme. The accuracy of the obtained models was evaluated using the root mean square error (RMSE) and the coefficient of determination (R²). The statistical analyses were carried out using XLSTAT statistical software (trial version).

3. Results and Discussion

3.1. Chemical Composition of Raw Meats and Meatballs and Cooking Measurements

In

Table 2. Chemical composition of raw meats and meatballs.

Meat Type	Pork Tenderloin			Turkey Tenderloin			Beef Tenderloin
Thermal treatment type	Raw	Steaming	Baking	Raw	Steaming	Baking	Raw	Steaming	Baking
Moisture, %	70.08 ± 0.22	52.48 ± 0.16	48.97 ± 0.12	67.29 ± 0.11	54.50 ± 0.09	50.44 ± 0.09	66.08 ± 0.18	52.86 ± 0.11	46.25 ± 0.11
Fat, g/100 g	3.50 ± 0.05	2.88 ± 0.06	2.68 ± 0.06	2.11 ± 0.08	2.00 ± 0.02	1.67 ± 0.04	7.00 ± 0.05	6.68 ± 0.03	5.85 ± 0.05
Protein, g/100 g	25.60 ± 0.20	24.50 ± 0.10	23.30 ± 0.22	30.60 ± 0.14	29.00 ± 0.12	28.20 ± 0.15	30.90 ± 0.03	30.00 ± 0.17	28.30 ± 0.16
Ash, g/100 g	0.98 ± 0.03	0.89 ± 0.02	1.1 ± 0.03	1.0 ± 0.04	1.1 ± 0.02	1.2 ± 0.03	1.35 ± 0.04	1.56 ± 0.02	1.75 ± 0.16

are presented the chemical properties of the meatballs before and after the thermal treatments. While no significant differences (p > 0.05) were observed between the samples with different types of herbal extract additions, only the results depending on the thermal treatments are presented in the table below.

As was expected, different cooking methods lead to different effects on the composition of the meatballs. Table 2 shows the main differences in the meatballs’ macroelements composition when using steaming and baking.

The differences in the moisture content caused by the cooking temperatures were significant (p < 0.05). The highest moisture content was determined for raw meats, varying from 66.08 ± 0.18 for beef tenderloin to 70.08 ± 0.22 for pork tenderloin.

Heat emanating from cooking is generally related to the structural and compositional denaturation of proteins [13]. As a proper explanation, water binding and tissue hydration processes happen in all the cellular disruption processes, especially sarcolemma, which determine the release of myofibrils which cause the hydration of the myofilament. Moreover, the denaturation of proteins which occurred during steaming and baking, as presented in Table 2, could be explained by the decrease in the sarcoplasmic hydrolysis rate by using a high temperature.

Fat loss is related to water evaporation as well as the ash content, which increased. Similar results were reported by Turhan et al. [14] in a study of meatballs formulated with different levels of bee pollen.

When talking about thermal treatments’ impact on the meatball’s moisture content, baking induced the highest decrease in the moisture content of the meatball samples, independent of the type of meat used. This could be associated with the oven’s inside conditions, especially related to the forced convection. This fact is also supported by the results obtained for the cooking loss and cooking yield.

• Cooking measurements

For each type of meatball, the mass (g), diameter (cm), core temperature (°C), and the dry matter content (%) were determined, as well as the textural characteristics every 3 min (hot air convection processed meatballs) or 2 min (water vapor convection processed meatballs), in order to carry out this study involving the influence of the heat transfer on the textural properties of the products. With the help of these determinations, the cooking loss and yield, the reduction in the volume, and the water retention of the meat samples were calculated, according to the equations presented in the Materials and Methods, and are presented in

Table 3. Cooking measurements of different types of meatballs, obtained by hot air convection and water vapor convection.

Analysis/Samples	ECPC	ERPC	ECPA	ERPA	ECCC	ERCC	ECCA	ERCA	ECVC	ERVC	ECVA	ERVA
Cooking loss, %	32.24 ± 0.51 B	38.56 ± 0.72 A	29.88 ± 0.64 C	28.10 ± 0.33 C,D	28.15 ± 0.52 C,D	29.04 ± 1.12 C,D	18.64 ± 0.86 E	18.79 ± 0.91 E	27.36 ± 0.78 D	27.75 ± 0.44 D	38.79 ± 0.25 A	38.03 ± 0.32 A
Cooking yield, %	67.76 ± 0.51 D	61.44 ± 0.72 E	70.12 ± 0.64 C	71.90 ± 0.33 B,C	71.85 ± 0.52 B,C	70.96 ± 1.12 B,C	81.36 ± 0.86 A	81.21 ± 0.91 A	72.64 ± 0.78 B	72.25 ± 0.44 B,C	61.21 ± 0.25 E	61.97 ± 0.32 E
Reduction in volume, %	11.43 ± 0.32 A	11.43 ± 0.41 A	9.09 ± 0.40 B	8.82 ± 0.30 B	8.57 ± 0.52 B	8.57 ± 0.58 B	8.57 ± 0.65 B	6.06 ± 0.25 C	8.57 ± 0.35 B	11.43 ± 0.45 A	8.82 ± 0.55 B	8.82 ± 0.78 B
Water retention, %	37.62 ± 0.42 C	33.18 ± 0.33 D	42.72 ± 0.45 B	44.50 ± 0.85 B	44.38 ± 0.64 B	44.36 ± 0.73 B	52.17 ± 0.56 A	51.41 ± 0.36 A	43.64 ± 0.94 B	43.49 ± 0.81 B	36.84 ± 0.47 C	37.04 ± 0.43 C
Dry matter, %	44.48 ± 0.42 A	46 ± 0.33 A	39.08 ± 0.45 B,C,D,E	38.11 ± 0.85 D,E,F	38.23 ± 0.64 C,D,E,F	37.48 ± 0.73 E,F,G	35.87 ± 0.56 G	36.7 ± 0.36 F,G	39.93 ± 0.94 B,C	39.81 ± 0.81 B,C,D	39.82 ± 0.47 B,C,D	40.22 ± 0.43 B

The averages on the same line with different superscripts are statistically significantly different (p < 0.05). ECPC and ERPC are baked pork meatballs with added wild thyme/lemon balm; ECPA and ERPA are steamed pork meatballs with added wild thyme/lemon balm; ECCC and ERCC are baked turkey meatballs with added wild thyme/lemon balm, while ECCA and ERCA are steamed turkey meatballs with added wild thyme/lemon balm. ECVC and ERVC are baked beef meatballs with added wild thyme/lemon balm, while ECVA and ERVA are steamed beef meatballs with added wild thyme/lemon balm.

.

The dry matter content of the aqueous-extract-enriched meatballs recorded values between 35.87% (ECCA) and 46% (ERPC). Comparable values of the dry matter content were obtained by Niu et al. in a study conducted on pork meatballs with the addition of gelatin and soluble dietary fiber from black beans, which could be directly linked to the type of thermal processing [15].

The reduction in the volume is a parameter related to water retention, being inversely proportional. As expected, the volume of myofibril is reduced by the thermal treatments, especially by the steaming of the turkey meatballs (6.06 ± 0.25%).

Water retention in meatballs could implicitly be associated with native hydrophilic colloids, widely employed in meatballs under certain conditions [5].

By analyzing the data in

Table 4. Influence of thermal treatments on aqueous-extract-enriched meatballs’ textural parameters.

Samples Parameters	ERPC	ECPC	ECPA	ERPA	ECCC	ERCC	ECCA	ERCA	ECVC	ERVC	ECVA	ERVA
Firmness, N	5.02 ± 0.42 A,B,C,D	4.50 ± 0.26 A,B	5.00 ± 0.32 A,B	5.58 ± 0.45 A	3.41 ± 0.79 C,D	5.01 ± 0.31 A,B	4.78 ± 0.52 A,B,C	3.10 ± 0.68 D	3.87 ± 0.14 B,C,D	3.98 ± 0.43 B,C,D	3.79 ± 0.89 B,C,D	5.26 ± 0.49 A,B
Adhesiveness, mJ	0.56 ± 0.08 A,B	0.47 ± 0.08 A	0.23 ± 0.03 D	0.26 ± 0.09 C,D	0.25 ± 0.02 D	0.32 ± 0.02 B,C,D	0.52 ± 0.07 A	0.33 ± 0.05 B,C,D	0.42 ± 0.05 A,B,C	0.29 ± 0.05 C,D	0.26 ± 0.03 C,D	0.55 ± 0.03 A
Cohesiveness	0.45 ± 0.03 A	0.52 ± 0.01 A,B	0.48 ± 0.05 A,B	0.48 ± 0.02 A,B	0.44 ± 0.04 A,B,C	0.39 ± 0.02 B,C	0.33 ± 0.03 C	0.44 ± 0.06 A,B,C	0.52 ± 0.02 A	0.50 ± 0.07 A,B	0.43 ± 0.03 A,B,C	0.44 ± 0.03 A,B,C
Springiness, mm	3.58 ± 0.44 A	3.72 ± 0.33 A	3.88 ± 0.19 A	3.81 ± 0.08 A	3.32 ± 0.25 A	3.04 ± 0.17 A	3.62 ± 0.26 A	3.73 ± 0.54 A	3.86 ± 0.62 A	3.81 ± 0.29 A	3.38 ± 0.28 A	3.75 ± 0.15 A
Gumminess, N	2.25 ± 0.52 A,B,C	2.34 ± 0.40 A,B,C,D	2.42 ± 0.27 A,B	2.66 ± 0.19 A	1.46 ± 0.18 C,D,E	1.95 ± 0.20 A,B,C,D,E	1.17 ± 0.25 E	1.39 ± 0.16 D,E	1.98 ± 0.37 A,B,C,D,E	2.01 ± 0.36 A,B,C,D,E	1.64 ± 0.14 B,C,D,E	2.31 ± 0.34 A,B,C
Chewiness, mJ	8.18 ± 1.61 A,B	8.8 ± 1.41 A, B,C,D	9.40 ± 1.37 A,B	10.15 ± 0.55 A	4.84 ± 0.47 D,E	5.93 ± 0.89 B,C,D,E	4.64 ± 0.38 E	5.31 ± 0.57 C,D,E	7.79 ± 1.62 A,B,C,D,E	7.69 ± 1.72 A,B,C,D,E	7.51 ± 0.83 A,B,C,D,E	8.66 ± 1.52 A,B,C

The averages on the same line with different superscripts (^A, ^B, ^C, ^D, and ^E) are statistically significantly different (p < 0.05). ECPC and ERPC are baked pork meatballs with added wild thyme/lemon balm; ECPA and ERPA are steamed pork meatballs with added wild thyme/lemon balm; ECCC and ERCC are baked turkey meatballs with added wild thyme/lemon balm, while ECCA and ERCA are steamed turkey meatballs with added wild thyme/lemon balm. ECVC and ERVC are baked beef meatballs with added wild thyme/lemon balm, while ECVA and ERVA are steamed beef meatballs with added wild thyme/lemon balm.

, it could be concluded that the textural parameters depend mostly on the type of meat and on the applied thermal treatment, so aqueous extracts added in the minced meat matrix did not change the specific texture profile of these kinds of gastronomic dishes.

The cooking loss recorded values between (ECVC) 27.36 ± 0.78 and 38.56 ± 0.72 (ERPC)% in the case of hot air convection processing and values between (ECCA) 18.64 ± 0.86 and 38.79 ± 0.25 (ECVA)% in the case of convection with water vapor processing. The thermal denaturation of meat proteins could be identified with changes in the structural features and water status in meat [16]. This phenomenon is highly correlated with the decrease in the acidic and basic groups of proteins, which shift the pH value and imply the isoelectric point of muscle proteins, inducing the decrease in the water retention capacity [17].

Inversely proportional to the cooking loss is the cooking yield, which recorded values between 61.44 (ERPC) and 72.64 (ECVC)% in the case of baking and values between 61.21 (ECVA) and 81.36 (ECCA) % in steam cooking. Similar cooking yield results were obtained by Brazhnaia et al. [18] for beef cutlets and by Lonergan et al. [19] for ground beef. A study on minced pork products with a low-fat content and the addition of Laminaria japonica powder, cooked using an electric grill, conducted by [20], reported values of the cooking loss between 13.78 and 34.64%. They stated that higher losses were recorded in the samples with a higher initial water content. Therefore, the fact that the meatballs with the addition of an aqueous extract of lemon balm/wild thyme had cooking losses 1.1–1.3× times higher compared to those reported by Choi et al. [20] may be due to the addition of an aqueous extract and also due to the difference in the water content compared to the raw material. For the same samples of minced meat with a low-fat content and addition of L. japonica powder, values of 10.04—18.84% were reported for the reduction in the volume. Compared to those mentioned above, the aqueous-extract-enriched meatballs recorded values of a reduction in the volume 1.6× times lower.

3.2. Texture Profile Analysis

During the thermic processing of meat, major structural changes take place (such as protein denaturation and collagen solubilization), which influence the textural characteristics of the finished product [21]. Using the Brookfield CT3 textural analyzer, the following parameters of the meatballs were determined: the firmness, adhesiveness, cohesiveness, springiness, gumminess, and chewiness. These parameters were chosen to be evaluated to correlate them with the sensory analysis.

In Table 4, the influence of thermal treatments on the aqueous-extract-enriched meatballs’ textural parameters is presented. According to the statement of Pathare and Roskilly, heat solubilizes the collagen which leads to fragility, although it distorts the myofibrillar proteins that lead to hardening [1]. All these heat-induced changes are time- and temperature-dependent, and the effect of this hardening or brittleness depends on the processing conditions. Minced pork meat samples achieved firmness values between 4.50 ± 0.26 N (ECPC) and 5.58 ± 0.45 N (ERPA), while samples of minced turkey meat had recorded values between 3.10 ± 0.68 N (ERCA) and 8.78 ± 0.52 N (ECCA), and the samples of minced beef, values between 3.79 ± 0.89 N (ECVA) and 5.26 ± 0.49 N (ERVA). In a study on minced pork products with the addition of gelatin and soluble dietary fiber from black beans, Niu et al. reported firmness values between 5.83 N and 10.8 N for the analyzed samples [15].

For both heat treatments applied, minced pork samples recorded higher values (8.18 ± 1.61 mJ–10.15 ± 0.55 mJ) of chewiness compared to the samples of minced turkey meat (4.64 ± 0.38 mJ–5.93 ± 0.89 mJ) and minced beef meat (7.51 ± 0.83 mJ–8.66 ± 1.52 mJ). These results show that pork meatballs require the highest amount of energy during the mastication (chewing) process. Studies on different types of meat microstructures revealed that the pork meat gains a more rigid and compact microstructure during cooking compared to beef meat [22], which could explain our results.

Instrumental cohesiveness indicates the strength of the internal bond in the product and is calculated as the ratio between the positive force area in the second compression and the first [23]. The lowest values of cohesiveness (0.33 ± 0.03–0.44 ± 0.4) were registered for the turkey samples and they could be associated with the values of chewiness. The cohesiveness values are comparable with those presented by Mabrouki et al. for homogenized meat and plant-based patties [24], Souppez et al. for beef burgers and plant-based analogues [25], and by Liu et al. for roasted duck breast [26].

In terms of gumminess, the samples of minced pork meat recorded slightly higher values (2.25 ± 0.52 N–2.66 ± 0.19 N) compared to the other samples of minced turkey and beef meat, which recorded values below 2 N for gumminess, both by baking and steam cooking. Niu et al. reported higher values (3.3 N–7.84 N) of the gumminess of minced pork products with the addition of gelatin and soluble dietary fiber from black beans, values that could be attributed to the addition of gelatin used [15].

Aqueous extracts added in the minced meat matrix did not change the specific texture profile of these kinds of gastronomic dishes.

3.3. Rheological Analysis

Aqueous-extract-enriched meatballs were subjected to a creep–recovery test to determine the influence of the heat treatments (convection with hot air and convection with water vapor) on their structure (Figure 1).

The creep–recovery test was performed in two successive distinct phases. Firstly, the creep test, which lasted for 295.6 s, followed by the recovery test, which was performed for 591.3 s. The upward curve represents the response of the samples to the creep test, while the downward part represents the ability to recover the initial shape of the sample. The appearance of the graphical representations is similar for all samples; meanwhile, the differences in values for the two parameters (creep and recovery) are influenced by the heat treatment. Thus, it can be observed that by using convection with water vapor processing, the demand for the samples (0.0009–0.0027) is influenced to a greater extent than by using the convection treatment with hot air (0.0009–0.0020), which means that the samples are more rigid. The values of the sample request are also influenced by the raw material used. As expected, for the products of minced pork meat (0.0009–0.0012) and those of minced turkey meat (0.0009–0.0014), lower requirements were needed compared to the samples of minced beef (0.0013–0.0027), which indicates a denser and stiffer structure of the minced beef samples.

From a rheological point of view, the samples exhibited a similar behavior, with the small differences in values being influenced either by the type of meat or thermal treatment which was expected due to the different composition and microstructure of the proteins [22].

A rheological model and its parameters are selected based on their adaptation to the experimental data. The viscoelastic behavior of meat products is generally described by using the generalized Maxwell model, in the case of stress–relaxation tests, while the Burger model usually supports the description of viscoelastic properties in creep tests [27]. Huang et al. [28] reported, in a study on the rheological properties, obtained by creep–recovery analysis, of myofibrillar protein from grass carp muscle (mowing), folding the request data to the Burger model. They suggested that this Burger model, together with the correlation coefficient (R²), is sufficient to describe the viscoelastic property and to reflect the internal structure of the myofibrillar protein samples of the mower.

Minced meat products with the addition of aqueous lemon balm/wild thyme extract do not fold on the Burger model, which uses exponential trend lines; instead, they fold on polynomial trend lines. This might be owed to the differences in the generally accepted assumptions while modeling the rheological data [29].

Neither the aqueous lemon balm extract, nor the wild thyme extract could be directly responsible for the rheological behavior of the meatball samples.

3.4. Differential Scanning Calorimetry (DSC)

Differential scanning calorimetry can be used to monitor protein denaturation. DSC studies performed on meat determined three stages of the denaturation of proteins influenced by the application of heat treatments, as classified by [30]. It begins with the denaturation of myosin at 40–60 °C, followed by the denaturation of sarcoplasmic proteins and collagen at 60–70 °C, and ending with actin denaturation at 80 °C.

From Figure 2, it can be noticed that all samples suffer mass losses in the temperature range of 2–8 °C due to the thawing process. According to Zhang and Ertbjerg, the loss due to thawing refers to the loss of water resulting from the formation of exudates after freezing and thawing. Such losses are lower after fast freezing compared to slow freezing [31].

Loss resulting from processing is a critical factor in the meat industry as it influences the technological efficiency of processing. From a nutritional perspective, a loss in processing involves the loss of soluble proteins and vitamins [32]. Devi et al. classified the processing losses caused by three temperature ranges and associated them with some structural components of the meat [33]. Thus, myosin denaturation corresponds to the range 54–58 °C, the change in the structure of collagen and sarcoplasmic proteins occurs at 65–67 °C, while the change in the structure of actin takes place at 80–83 °C. They used differential scanning calorimetry on camel muscles and reported processing losses at 58.17 °C, 68.48 °C, and 84.16 °C, which were attributed to the aforementioned phenomena. Purslow et al. also argue that the meat losses suffered after processing all types of meat are mainly caused by thermal denaturation [32]. At temperatures above 42 °C, myosin is denatured by producing the lateral contraction of muscle fibers, while at higher temperatures (70–80 °C), actin is denatured due to the longitudinal contraction of muscle fibers. In addition, it was reported that myosin in white fibers is less thermally stable and, therefore, more susceptible to denaturation than myosin in red fibers in both beef and chicken.

3.5. Sensory Evaluation

According to [1], the processing methods and raw material quality change the sensory characteristics of the product. Among the reliable and consistent measures for assessing the characteristics of meat is the sensory evaluation. Thus, the acceptability of the analyzed product largely depends on the decision of the final consumer.

The chi-squared test has been used to verify the normality of the selected values. Two hypotheses were assumed:

H0: 

The sample follows a normal distribution;

Ha: 

The sample does not follow a normal distribution.

The histogram of the results is presented in Figure 3.

The calculated chi-square values are between 1.7588 and 10.3628 and they are lower compared with the critical chi-square value (14.0671). As our computed p-values are greater than the significance level alpha = 0.05 and the values of the calculated chi-square are below the critical values, the alternative hypothesis Ha is rejected and the null hypothesis H0 (that the data follow a normal distribution) is accepted.

Table 5. Pearson’s correlations values of the sensory characteristics, textural parameters, and cooking measurements of aqueous-extract-enriched meatballs.

Variables	CL	CY	RV	WR	DM	F	C	S	G	CW	A	T	AM	AT	MF	F1	E	C2	J	OA
CL	1	−1.0000	0.5142	−0.9701	0.7053	0.2664	0.3843	−0.0981	0.5478	0.5849	0.1784	0.1347	0.3618	0.1625	−0.0715	0.2541	0.0332	−0.1129	−0.7046	−0.2040
CY	−1.0000	1	−0.5142	0.9701	−0.7053	−0.2664	−0.3843	0.0981	−0.5478	−0.5849	−0.1784	−0.1347	−0.3618	−0.1625	0.0715	−0.2541	−0.0332	0.1129	0.7046	0.2040
RV	0.5142	−0.5142	1	−0.6439	0.7742	0.3302	0.3966	0.1267	0.4875	0.4567	0.1816	0.1194	0.2968	−0.0570	−0.0281	0.0691	−0.2274	−0.3097	−0.7128	−0.0827
WR	−0.9701	0.9701	−0.6439	1	−0.8551	−0.2500	−0.4874	0.0088	−0.5905	−0.6153	−0.1586	−0.1149	−0.3565	−0.0613	0.1339	−0.1423	0.0986	0.1190	0.7685	0.2564
DM	0.7053	−0.7053	0.7742	−0.8551	1	0.1601	0.5777	0.1788	0.5325	0.5312	0.0402	0.0158	0.2532	−0.2091	−0.2786	−0.1542	−0.3822	−0.0918	−0.7228	−0.3521
F	0.2664	−0.2664	0.3302	−0.2500	0.1601	1	−0.1142	0.0962	0.6733	0.5703	−0.0468	0.0095	−0.0029	0.1596	0.1555	0.3706	−0.2600	−0.1958	−0.2693	−0.0111
C	0.3843	−0.3843	0.3966	−0.4874	0.5777	−0.1142	1	0.5651	0.6269	0.6689	0.4038	0.3994	0.4912	0.3687	0.4123	0.3135	0.0681	−0.2072	−0.5967	0.0919
S	−0.0981	0.0981	0.1267	0.0088	0.1788	0.0962	0.5651	1	0.3898	0.5497	0.4254	0.3084	0.2506	0.1932	0.3526	0.4857	0.1491	−0.0952	−0.4414	0.0906
G	0.5478	−0.5478	0.4875	−0.5905	0.5325	0.6733	0.6269	0.3898	1	0.9356	0.2220	0.2686	0.3392	0.4480	0.4012	0.4994	−0.1528	−0.2150	−0.6074	0.0295
CW	0.5849	−0.5849	0.4567	−0.6153	0.5312	0.5703	0.6689	0.5497	0.9356	1	0.3501	0.3491	0.4115	0.4870	0.4337	0.6101	−0.0377	−0.2238	−0.7094	0.0518
A	0.1784	−0.1784	0.1816	−0.1586	0.0402	−0.0468	0.4038	0.4254	0.2220	0.3501	1	0.9061	0.8151	0.6945	0.6538	0.6204	0.8017	0.1718	−0.6406	0.7527
T	0.1347	−0.1347	0.1194	−0.1149	0.0158	0.0095	0.3994	0.3084	0.2686	0.3491	0.9061	1	0.9114	0.7916	0.7369	0.5029	0.6946	0.2948	−0.5201	0.8913
AM	0.3618	−0.3618	0.2968	−0.3565	0.2532	−0.0029	0.4912	0.2506	0.3392	0.4115	0.8151	0.9114	1	0.6936	0.6562	0.3995	0.6174	0.3325	−0.6000	0.6932
AT	0.1625	−0.1625	−0.0570	−0.0613	−0.2091	0.1596	0.3687	0.1932	0.4480	0.4870	0.6945	0.7916	0.6936	1	0.8785	0.7197	0.6335	0.1465	−0.3013	0.7695
MF	−0.0715	0.0715	−0.0281	0.1339	−0.2786	0.1555	0.4123	0.3526	0.4012	0.4337	0.6538	0.7369	0.6562	0.8785	1	0.6956	0.5830	−0.0648	−0.1855	0.7472
F1	0.2541	−0.2541	0.0691	−0.1423	−0.1542	0.3706	0.3135	0.4857	0.4994	0.6101	0.6204	0.5029	0.3995	0.7197	0.6956	1	0.6007	−0.3005	−0.4665	0.4172
E	0.0332	−0.0332	−0.2274	0.0986	−0.3822	−0.2600	0.0681	0.1491	−0.1528	−0.0377	0.8017	0.6946	0.6174	0.6335	0.5830	0.6007	1	0.2201	−0.2279	0.6696
C2	−0.1129	0.1129	−0.3097	0.1190	−0.0918	−0.1958	−0.2072	−0.0952	−0.2150	−0.2238	0.1718	0.2948	0.3325	0.1465	−0.0648	−0.3005	0.2201	1	0.1017	0.2588
J	−0.7046	0.7046	−0.7128	0.7685	−0.7228	−0.2693	−0.5967	−0.4414	−0.6074	−0.7094	−0.6406	−0.5201	−0.6000	−0.3013	−0.1855	−0.4665	−0.2279	0.1017	1	−0.1647
OA	−0.2040	0.2040	−0.0827	0.2564	−0.3521	−0.0111	0.0919	0.0906	0.0295	0.0518	0.7527	0.8913	0.6932	0.7695	0.7472	0.4172	0.6696	0.2588	−0.1647	1

CL, cooking loss; CY, cooking yield; RV, reduction in volume; WR, water retention; DM, dry matter; F, firmness; C, cohesiveness; S, springiness; G, gumminess; CW, chewiness; A, appearance; T, taste; AM, aroma; AT, aftertaste; MF, mouthfeel; F₁, firmness (sensorial); E, elasticity; C₂, cohesiveness (sensorial); J, juiciness; OA, overall acceptability. Values in bold are different from 0 with a significance level alpha = 0.05.

presents the correlation matrix between the quality characteristics tested. Several characteristics were found to be greatly correlated with each other. The highest significant positive and negative correlations were recorded between the cooking loss and water retention, cooking yield and water retention, and taste and aroma. This aspect was expected because when the value of the cooking loss is high, more water is excluded from the meatballs, leading to lower water retention.

The addition of the aqueous extract of lemon balm/wild thyme, along with processing methods, influences the taste and aroma of meatballs, making them more enjoyable for panellists. The results of the sensory evaluation of meatballs with an aqueous extract of lemon balm/wild thyme are presented in

Table 6. Sensory analysis results of cooked meatballs with aqueous extract of lemon balm/wild thyme.

Samples/Sensorial Characteristic	ECPC	ERPC	ECPA	ERPA	ECCC	ERCC	ECCA	ERCA	ECVC	ERVC	ECVA	ERVA
Appearance	7.0 ± 1.49 C,D	6.5 ± 1.19 C,D	7.1 ± 1.45 C,D	7.1 ± 1.37 C,D	6.4 ± 1.63 D	6.8 ± 1.57 C,D	6.7 ± 1.11 C,D	6.8 ± 1.30 D	8.4 ± 0.70 A,B	8.6 ± 0.52 A	7.6 ± 1.07 B,C,D	8.0 ± 0.67 A,B,C
Taste	7.2 ± 1.55 A,B,C	7.2 ± 1.55 B,C	7.1 ± 1.45 B,C	7.6 ± 0.70 A,B,C	6.8 ± 1.48 C	7.5 ± 1.27 A,B,C	7.1 ± 1.20 A,B,C	7.3 ± 1.42 A,B,C	8.1 ± 0.88 A	8.1 ± 0.74 A	7.6 ± 0.97 A,B,C	7.7 ± 1.06 A,B
Aroma	7.1 ± 1.79 A,B,C	7.6 ± 1.71 A,B,C	7.2 ± 1.40 B,C	7.3 ± 0.95 A,B,C	6.6 ± 1.58 C	7.6 ± 1.78 A,B	6.9 ± 1.20 A,B,C	7.2 ± 1.48 B,C	8.0 ± 1.05 A,B	8.2 ± 0.79 A	7.7 ± 0.67 A,B,C	7.7 ± 0.82 A,B
Aftertaste	7.1 ± 1.29 A	7.1 ± 1.79 A	7.4 ± 0.84 A	7.7 ± 1.16 A	7.1 ± 1.60 A	7.5 ± 1.35 A	7.0 ± 1.25 A	7.3 ± 1.57 A	7.5 ± 0.71 A	7.7 ± 0.82 A	7.5 ± 1.18 A	7.5 ± 1.08 A
Mouthfeel	6.9 ± 1.52 A	7.1 ± 1.73 A	7.6 ± 0.97 A	7.7 ± 0.82 A	7.0 ± 1.15 A	7.5 ± 1.08 A	7.1 ± 1.37 A	7.2 ± 1.40 A	7.7 ± 0.67 A	7.8 ± 0.92 A	7.3 ± 0.95 A	7.3 ± 0.95 A
Firmness	7.2 ± 1.69 A	7.3 ± 1.77 A	7.8 ± 0.92 A	7.9 ± 0.74 A	7.4 ± 0.97 A	7.3 ± 1.06 A	7.4 ± 1.07 A	7.2 ± 1.03 A	7.6 ± 0.84 A	7.8 ± 1.03 A	7.6 ± 0.70 A	7.9 ± 0.88 A
Cohesiveness	6.6 ± 1.51 A	6.5 ± 1.90 A	6.4 ± 1.78 A	6.4 ± 1.84 A	6.3 ± 1.42 A	6.7 ± 1.95 A	6.5 ± 1.78 A	6.9 ± 1.85 A	6.5 ± 1.51 A	6.6 ± 1.26 A	6.6 ± 1.18 A	6.7 ± 1.16 A
Juiciness	4.9 ± 0.76 B	5.0 ± 0.86 A,B	5.6 ± 0.85 A,B	5.5 ± 0.92 A,B	6.1 ± 0.79 A,B	6.3 ± 0.89 A	6.2 ± 0.98 A,B	6.5 ± 0.97 A,B	5.1 ± 0.96 A,B	4.8 ± 0.78 A,B	5.2 ± 0.94 A,B	4.7 ± 1.01 B
Overal acceptability	7.2 ± 1.81 A,B	7.0 ± 1.70 B	7.1 ± 1.37 A,B	7.7 ± 1.48 A,B	7.1 ± 0.99 A,B	7.7 ± 1.06 A,B	7.4 ± 0.97 A,B	7.4 ± 1.35 A,B	7.9 ± 0.74 A	8.0 ± 0.67 A,B	7.5 ± 0.53 A,B	7.7 ± 0.67 A,B

The averages on the same line with different superscripts (^A, ^B, ^C, and ^D) are statistically significantly different (p < 0.05). ECPC and ERPC are baked pork meatballs with added wild thyme/lemon balm; ECPA and ERPA are steamed pork meatballs with added wild thyme/lemon balm; ECCC and ERCC are baked turkey meatballs with added wild thyme/lemon balm, while ECCA and ERCA are steamed turkey meatballs with added wild thyme/lemon balm. ECVC and ERVC are baked beef meatballs with added wild thyme/lemon balm, while ECVA and ERVA are steamed beef meatballs with added wild thyme/lemon balm.

.

Sensorial characteristics such as the appearance, taste, and aftertaste were positively correlated with the overall acceptance of the meatballs. The juiciness was negatively correlated to the cooking loss; a high cooking loss results in low juiciness. Similar results were reported by Aaslyng et al. in their study about the influence of the raw meat quality and type of cooking process on the cooking loss and juiciness of pork meat [34].

The textural parameters and sensory characteristics were subjected to PLS regression with the goal to assess a prediction model for sensory attributes. The values of the coefficient of determination were between 0.9938 and 0.9997, while the values of the root mean square error ranged between 0.0028 and 0.0286. Based on the obtained results, combining the textural parameters with sensory characteristics can provide a more complete understanding between the product characteristics (meatballs with an aqueous extract of lemon balm/wild thyme) and the panellists’ perception.

In PCA, the first two PCs were sufficient to explain the maximum variation for the meatball’s sensorial characteristics, textural parameters, and cooking measurements (

Table 7. Explained variance.

	Eigenvalue	Variability, %	Cumulative, %
PC1	7.9135	39.5675	39.5675
PC2	5.3116	26.5578	66.1254
PC3	2.1443	10.7217	76.8471
PC4	1.4712	7.3561	84.2031
PC5	1.0766	5.3829	89.5860
PC6	0.8574	4.2871	93.8732
PC7	0.7559	3.7793	97.6525
PC8	0.1896	0.9479	98.6004
PC9	0.1497	0.7486	99.3490
PC10	0.0845	0.4225	99.7715
PC11	0.0457	0.2285	100

). Mwove et al. identified three PCs explaining 73.63% of the total variation for the physicochemical, textural, and sensorial attributes of beef rounds with gum arabic [35]. The PCs’ loadings (coefficients of the correlation between the variable and PCs) for the meatballs’ sensory characteristics, cooking measurements, and textural parameters are presented in

Table 8. The PCs’ loadings (coefficients of correlation between variable and PCs).

	PC 1	PC 2	PC 3	PC 4	PC 5	PC 6
Cooking loss	−0.6337	−0.5847	0.3046	−0.3209	−0.1942	−0.1432
Cooking yield	0.6337	0.5847	−0.3046	0.3209	0.1942	0.1432
Reduction in volume	−0.5191	−0.5492	0.0658	0.1454	0.1155	0.5610
Water retention	0.6397	0.6812	−0.3035	0.1361	0.0909	0.0816
Dry matter	−0.4876	−0.7677	0.2499	0.2846	0.1544	0.0522
Firmness	−0.3260	−0.2783	−0.5430	−0.5098	0.3940	0.2010
Cohesiveness	−0.6988	−0.1550	−0.0812	0.5375	−0.0952	−0.2539
Springiness	−0.4497	0.1200	−0.4393	0.6367	0.0021	−0.1149
Gumminess	−0.7536	−0.3448	−0.4281	−0.0983	0.2492	−0.1616
Chewiness	−0.8292	−0.2900	−0.3951	−0.0036	0.1124	−0.1978
Appearance	−0.7466	0.5272	0.2127	0.1299	−0.0915	0.1686
Taste	−0.7321	0.5852	0.2310	0.0493	0.1659	0.0974
Aroma	−0.7940	0.3582	0.3777	0.0518	0.1731	0.0256
Aftertaste	−0.6893	0.5788	−0.1000	−0.2575	0.0414	−0.1874
Mouthfeel	−0.6055	0.6415	−0.3106	−0.0468	0.0145	−0.0241
Firmness	−0.6840	0.3262	−0.4308	−0.2178	−0.3614	−0.0122
Elasticity	−0.4088	0.7230	0.2864	−0.1149	−0.3818	0.0014
Cohesiveness	0.0646	0.3226	0.5584	−0.0239	0.5766	−0.3702
Juiciness	0.8680	0.2915	−0.1599	−0.1446	0.0518	−0.2000
Overall acceptance	−0.4285	0.8003	0.1152	−0.0903	0.1934	0.2400

. The first PC (39.57%) was related to the textural parameters (cohesiveness, gumminess, and chewiness), sensorial characteristics (appearance, taste, aroma, aftertaste, firmness, and juiciness), and cooking measurements (cooking loss and cooking yield), while the second PC (26.56%) was associated with water retention, dry matter (cooking measurements), and sensorial parameters (mouthfeel, elasticity, and overall acceptability). As a consequence of the PCA, 11 principal component axes were acquired, and these axes constitute the total variation.

A graphical representation of the eigenvalues (Scree Plot) is presented in Figure 4. According to [36], if the eigenvalues are greater than 1, the evaluated principal component weight values are reliable.

Figure 5 shows the results obtained from Principal Component Analysis (PCA) of the sensorial analysis, textural analysis, and cooking measurements for the 12 meatball samples, in which the first two principal factors could explain 66.13% of the variance. The measurements and PCs are explained according to the correlations between each parameter and each PC. According to Mwove et al. [35], the measurements close together are positively correlated, separated at 180° if they are negatively correlated, or independent if they are separated by 90°. When analyzing the biplot graph of the meatball samples used in the present study (Figure 5), it is observed that the meatballs are different from each other and are spread out across the graph. All the meatballs’ samples are located between the first and the third quadrant. Similar results were observed also by Przybylski et al. in a study performed on meat products with bioactive compounds [37]. In the first quadrant, three meatballs’ samples (ERCC, ERCA, and ECCA) obtained from minced turkey tenderloin are presented. In the second quadrant, the meatballs samples ERVC, ECVC, and ERPA, obtained from minced beef tenderloin, are presented. The third quadrant consisted of steamed meatballs obtained from minced pork tenderloin (ERPC and ECPA) and minced beef tenderloin (ECVA and ERVA). In the fourth quadrant, there were only two meatball samples (ECPC and ECCC) treated by hot air convection (baking).

The results of the HCA were accordant with the results of the sensory evaluations, textural analysis, and cooking measurements, in which all meatball samples were distinctly divided into three groups (Figure 6). The first group contains samples obtained from minced pork tenderloin. The meatballs in group two are manufactured from minced beef tenderloin, while the third group includes samples obtained from minced turkey tenderloin. This distribution, presented in Figure 5, demonstrates that the meat composition and cooking methods have a great influence on the final products. The HCA results were similar to those of the PCA.

4. Conclusions

One factor that determines the texture of products made from minced meat is the impact of the heat treatment. In this case, steam cooking determined a firmer texture for all samples compared to those obtained by baking. In addition, the results of the textural determinations showed that the samples obtained from minced turkey meat processed by convection with hot air, but also those processed by convection with water vapor, required a smaller amount of energy during the chewing process compared to the other samples analysed.

The three meat types present specific endothermic transitions and protein denaturation predominantly dependent on the number of cross-links, as well as on the synergic action of other denaturing agents during DSC analysis.

By the technological meaning of the experiment, the wild thyme and lemon balm aqueous extracts, even recognized for their health benefits, do not have a distinguishable impact independent of the applied heat treatments. Neither in terms of the rheological nor textural properties do the addition of both aqueous extracts have a visible impact.

A further perspective of this study could consist in some other detailed type of rheological tests.