Abstract
A traditional dairy-based frozen dessert (ice cream) was developed with three levels of sweetpotato (Ipomoea batatas) puree [20%, 30%, and 40% (by weight)] to determine the impact of sweetpotato content on product functionality, nutritional content, and sensory characteristics. Increased sweetpotato puree resulted in increased orange color, flavor intensity, and sweetpotato flavor, but 40% puree proved difficult to incorporate into the mixture. Additionally, nondairy frozen desserts containing 30% sweetpotato puree were compared with a milk-based control in which all ingredients were the same except that milk was replaced with soy (Glycine max) and almond (Prunus dulcis) milk. Consumer acceptability tests were conducted with panelists at Mississippi State University (n = 101) and in Pontotoc, MS (n = 43). Panelists in Pontotoc rated the overall acceptability of all three frozen desserts the same, but they preferred the appearance of the milk-based frozen dessert over that of soy- and almond-based milk alternatives. According to the panelists at Mississippi State, the milk-based frozen dessert had greater overall acceptability and aroma than the almond-based dessert and a preferential texture and appearance compared with the soy- and almond-based desserts. Milk-, soy-, and almond-based frozen desserts were rated as “slightly liked” or better by 92%, 80%, and 69% of the panelists, respectively.
Sweetpotato (Ipomoea batatas) is the seventh most consumed staple food in the world and is considered a food security staple in developing countries due to its adaptability to climate change and productivity under drought conditions (Bonvell-Benjamin, 2007). The main component of sweetpotato is complex starch, yielding one of the densest caloric root vegetables, 86 kcal/100 g [U.S. Department of Agriculture (USDA), 2018]. Sweetpotato has a lower glycemic index (GI) than simple carbohydrate sources, and the abundance of fiber prevents the absorption of glucose in the small intestine, thereby reducing the rate of insulin secretion (Jenkins et al., 1988). For this reason, sweetpotato is considered a good substitute in diabetic and weight control diets for simple, starchy foods such as white bread and potatoes (Solanum tuberosum) (Jenkins et al., 1988). The crop also has a variety of nutritional benefits with respect to both macro- and micronutrients, is abundant in antioxidants, and orange-fleshed cultivars are prominent in carotenoids (Burri, 2011). United States sweetpotato production has greatly increased during the past 15 years, reaching 3.1 billion pounds in 2015 [USDA-National Agriculture Statistics Service (NASS), 2018]. Per capita consumption of sweetpotato increased from 4.2 to 7.6 lb from 2000 to 2015 due in part to awareness of its nutritive value and its increased use in value-added food products.
Mississippi is the third largest producer of sweetpotatoes after North Carolina and California (USDA-NASS, 2018). Sweetpotato is the fourth most valuable crop in Mississippi after soybeans (Glycine max), cotton (Gossypium hirsutum), and corn (Zea mays) (USDA-NASS, 2018). Between 2000 and 2015, Mississippi sweetpotato production increased by 155% (Johnson et al., 2015). In 2017, 29,000 acres were harvested (USDA-NASS, 2018). During an evaluation of the 2011 season, Morgan et al. (2012) reported that the Mississippi sweetpotato industry employed 1059 full-time equivalents. With the steady growth in recent years, the direct output has doubled from $66.4 million to $123 million in 2017 (USDA-NASS, 2018), with nearly an equivalent indirect impact (Morgan et al., 2012).
In addition to the increased production and consumption of sweetpotatoes, there is also increasing scrutiny of produce buyers and consumers of aesthetically pleasing sweetpotatoes meeting stringent guidelines for qualities, including size and shape. For example, Collart et al. (2019) reported that consumer willingness-to-pay for sweetpotatoes decreased as the portion of the sweetpotato skin affected by skinning injury increased. For the 2018 crop year, Meyers (2019) reported that many Mississippi producers did not harvest processing grade sweetpotatoes because of their low value and lack of value-added outlets. As a result, there is a need to develop additional value-added uses for processing grade sweetpotatoes that do not meet stringent standards required to be sold as a fresh market USDA no. 1 grade. One sector of value-added foods that has the potential for using processing grade sweetpotatoes is ice creams and nondairy frozen desserts.
Ice cream is a semi-solid foam or custard made from dairy products such as milk, cream, and nonfat milk powder. This complex and sweet-tasting colloid is frozen below its freezing point so that it is smooth and creamy (Goff, 1997). Commercially produced ice cream incorporates air for a desirable texture and increased volume. Ice cream is a standardized food in the United States that is defined as a dairy frozen dessert when there is a minimum of 10% fat and less than 50% overrun (U.S. Food and Drug Administration, 2018). Ice cream is categorized as economy, premium, and super premium based on its fat content, total solids, and overrun. The greater the amount of fat added and the less air incorporated will result in a smoother ice cream. However, dairy-based products, including ice cream, should not be consumed by those requiring a dairy-free diet because of lactose intolerance, dairy allergies, and/or veganism (Mäkinen et al., 2016).
The global nondairy frozen dessert market reached $400 million in 2017 (Ahuja and Rawat, 2018). Plant-based milk alternatives used in frozen desserts are typically opaque liquid extracted from legumes or tree nuts, most commonly almond (Prunus dulcis) and soybean. Compared with milk, plant-based milk alternatives are lower in calories and comparable in calcium content (Mäkinen et al., 2016). In nondairy frozen desserts, cream is replaced by vegetable oil to maintain a smooth texture; however, coconut (Cocos nucifera) oil can also be used to improve the texture of frozen desserts by functioning as an emulsifier (Santana et al., 2011). Coconut oil is a highly saturated oil with a sweet aroma. It is a natural source of medium-chain fatty acids (MCFAs), with more than 50% MCFAs. MCFAs consist of 6 to 10 carbons; therefore, the hydrolysis and absorption of MCFAs is more efficient than that of long-chain fatty acids because the mechanism of its digestion is similar to glucose (Marten et al., 2006). Coconut oil has a high melting point (24.4 °C), which enables the possibility of its use in frozen desserts. Additionally, coconut oil is used in many commercial products to improve health benefits (Choo et al., 2010) and shelf stability.
In nondairy frozen desserts, milk is replaced with fortified plant-based milk alternatives. Whole milk contains 3.3% protein and almond milk contains 0.5% protein, which is the lowest concentration of the 14 different milk alternatives on the market. The protein content of soy milk varies according to manufacturers. Almond milk and soy milk contain more fiber and less saturated fat than milk (Mäkinen et al., 2016). Although the market for nondairy frozen desserts is rapidly growing, they are not as acceptable as ice cream. Bisla et al. (2012) reported that ice cream made from 100% milk was preferred over that made from 100% soy and other plant-based milk alternative “ice creams.” In a sensory evaluation study of milk (whole, reduced fat, and fat-free) and soy milk (vanilla, fortified, and organic), milk was preferred (Palacios et al., 2009). The objective of this study was to develop sweetpotato-based ice creams and nondairy frozen desserts to add value to underused processing grade sweetpotatoes.
Materials and methods
Product development
Product development was conducted at Mississippi State University’s Ammerman-Hernsberger Food Processing Laboratory and the Edward W. Custer Dairy Processing Plant, Mississippi State, MS, and it occurred in two phases. During the first phase, a traditional dairy-based frozen dessert (ice cream) was developed with three levels of sweetpotato puree [20%, 30%, and 40% (by weight)] to determine the impact of sweetpotato content on product functionality, nutritional content, and sensory characteristics by a small, semi-trained panel. The optimum level of sweetpotato puree in the dairy-based frozen dessert from the first phase was used during the second phase to develop nondairy frozen dessert products that were evaluated by nontrained consumer participants.
‘Beauregard’ sweetpotatoes were secured from the Pontotoc Ridge-Flatwoods Branch Experiment Station, Pontotoc, MS, and stored at room temperature and 70% relative humidity in the Food Processing Laboratory for ≈4 months before being used. Milk, heavy cream, almond milk, nonfat milk solids, soymilk, coconut oil, cane sugar (extra fine, granulated), stabilizer (PGX-1; Germantown Manufacturing Co., Broomall, PA), and vanilla flavor were purchased from a local store. Sweetpotatoes were baked at 190 °C in a convection oven (HEC20; Hobart Corp., Troy, OH) for 60 min. After baking, the skin was removed and the pulp was cooled for 30 min. Then, the pulp was pureed at high speed in a food processor (FP-8SV; Cuisinart, Stamford, CT) for 3 min. The puree was stored at −18 °C and then thawed at 4 °C in a refrigerator for 24 h before use. Phase one dairy-based frozen desserts contained milk, heavy cream, sugar, nonfat milk solids, stabilizer, vanilla flavor, and 20%, 30%, or 40% sweetpotato puree (Table 1). The ingredient amounts were adjusted with an ice cream calculator to maintain the same percentages of fat (10%) and sugar (16%) in each sample. Based on findings from the dairy dessert sensory panel (see Results and Discussion), phase two nondairy frozen desserts contained 30% sweetpotato puree and used coconut oil as a replacement for heavy cream and either almond milk or soy milk as a replacement for milk (Table 2). Nondairy desserts were compared with a milk-based control that comprised all the same ingredients except that milk replaced the soy milk and almond milk. For both phase one and phase two, all ingredients except sweetpotato puree and vanilla were mixed and heated in a pot (diameter, 30 cm; height, 16 cm) until the mixture reached 50 °C. Then, the mixture was homogenized with sweetpotato puree and vanilla flavor in a food processor at the low setting for 60 s and cooled at 4 °C in a refrigerator for 18 h. An ice cream machine (BCI600XL; Breville USA, Torrance, CA) was used for initial freezing for 40 min, and the samples were stored in plastic containers for the hardening process at −40 °C until they were analyzed.
Formulations of sweetpotato frozen desserts made from milk and cream with different sweetpotato contents.
Formulations of sweetpotato frozen desserts made with milk, almond milk, or soy milk.
Chemical analysis
The AOAC methods 990.30 and 934.01 were used to determine the protein and moisture, respectively (Association of Official Agricultural Chemists International, 1999). Fiber was measured by the American Oil Chemists’ Society (AOCS) Ba6A-05 (American Oil Chemists’ Society, 1998). To determine minerals, inductively coupled plasma mass spectrometry (ICP-MS) was used for metal analysis. Then, 2 g of frozen dessert was weighed and transferred to a microwave digestion tube with 3 mL of peroxide and 5 mL of nitric acid (HNO3). The tube was capped and place in a carousel in a hood for 2 h. The carousel was placed in a microwave digester (MARS Xpress; CEM Corp., Matthews, NC). The tube was cooled for 12 h, the cap was removed, and the acid was equilibrated for 30 min in the hood. The digested samples were added to each volumetric flask with 50 mL of deionized water and filtered into the ICP sampler tubes with 0.45 μm and 33 mm polyvinylidene difluoride filters. The ICP-MS analyzer (7900 ICP-MS; Agilent, Santa Clara, CA) used gas plasma to determine the calcium, iron, sodium, and potassium levels.
Nutrition estimation analysis
The nutritional content of frozen desserts was estimated by a calorie calculator (Tufts University, 2019). According to the frozen dessert formula (Table 2), ingredients were searched in the USDA food composition database and calculated using the USDA software. Then, 100 g of frozen dessert was established as one serving.
Sensory analysis
During phase one, a descriptive sensory test of dairy-based frozen dessert formulations was conducted with a semi-trained panel (n = 8) in a descriptive room at the Mississippi State University Garrison Sensory Laboratory, Mississippi State, MS. Attributes (Table 3) were derived from those used by King (1994) and Ohmes et al. (1998); they were evaluated by the panelists using a scale of 0 (very weak) to 15 (very strong) with two replications. Interactions between panelists and treatments were analyzed by SAS (version 9.4; SAS Institute, Cary, NC). Data from two panelists were excluded to reduce outliers in the final statistical analysis (Lawless and Heymann, 2010).
Attributes and their definitions used by a semi-trained (n = 8) descriptive panel who evaluated frozen dairy desserts at the Mississippi State University Garrison Sensory Laboratory, Mississippi State, MS.
During phase two, consumer acceptability tests involving the nondairy-based frozen desserts and a milk-based control were conducted at the Mississippi State University Garrison Sensory Laboratory (n = 101) and at the 2018 Annual Sweetpotato Field Day, Pontotoc Ridge-Flatwoods Branch Experiment Station, Pontotoc, MS (n = 43). Participants at the sensory laboratory were university students or employees. Participants at the Field Day were sweetpotato growers and university or industry research/extension personnel who were generally more knowledgeable about sweetpotatoes than participants at Mississippi State University. A hedonic scale of 1 to 9 was used to rate the appearance, aroma, flavor, texture, and overall liking of each of the frozen dessert samples.
Statistical analysis
For descriptive analysis (phase one), a randomized complete block design was used. An interaction plot was used to exclude inconsistent panelists. The remaining descriptive and chemical results were subjected to an analysis of variance by SAS with Tukey’s honestly significant difference to differentiate the treatment means when P ≤ 0.05. For phase two, a randomized complete block design was used to evaluate differences (P ≤ 0.05) in the appearance, aroma, flavor, texture, and overall acceptability of sweetpotato frozen desserts. Panelists were treated as blocks. A cluster analysis was used for the 101 panelists who participated in the consumer acceptability test at Mississippi State University; they were clustered by dissimilarities in their overall liking and preference of the samples using Ward’s method of agglomerative hierarchy clustering with XLSTAT (Addinsoft, New York, NY). The number of clusters was determined by comparing the levels on a dissimilarity plot. Each cluster was analyzed by SAS, and Tukey’s test was used to separate the treatment means when a difference (P ≤ 0.05) occurred.
Results and discussion
Determining the optimum sweetpotato puree content in frozen dairy desserts (phase one)
Using the aforementioned scale of 0 (very weak) to 15 (very strong), the semi-trained sensory panel determined that the intensity of the orange color increased with the increasing sweetpotato puree content; the intensity scores were 7.1, 8.6, and 11.1 for dairy-based desserts containing 20%, 30%, and 40%, respectively (Fig. 1). The amount of sweetpotato added determines the intensity of the orange color as a result of the natural beta carotene pigment (Takahata et al., 1993). All samples received low ratings for aroma attributes (4.0–5.4), coarse iciness (2.9–3.5), gumminess (≤4.8), sourness (0.6–0.8), and astringency (0.9–1.2). The overall flavor intensity was stronger with 40% sweetpotato puree than with 20%; the flavor intensity with 30% was similar to that with 20% and 40%. However, the formulation containing 40% sweetpotato was more difficult to produce because of the difficulty incorporating both puree and air into the product. Furthermore, the authors had concerns about the stability of the 40% formulation product. Therefore, the 30% sweetpotato puree was chosen for the frozen dessert samples used during phase two.
Mean scores of descriptive panels for each attribute of frozen dairy desserts made with 20%, 30%, and 40% sweetpotato puree. A 15-cm (5.9 inches) line scale was used for each attribute: 0 = very weak and 15 = very strong. Tukey’s honestly significant difference test was used to separate treatment means (P ≤ 0.05). Means for each attribute followed by the same letter do not differ.
Citation: HortTechnology hortte 31, 3; 10.21273/HORTTECH04725-20
Mean scores of descriptive panels for each attribute of frozen dairy desserts made with 20%, 30%, and 40% sweetpotato puree. A 15-cm (5.9 inches) line scale was used for each attribute: 0 = very weak and 15 = very strong. Tukey’s honestly significant difference test was used to separate treatment means (P ≤ 0.05). Means for each attribute followed by the same letter do not differ.
Citation: HortTechnology hortte 31, 3; 10.21273/HORTTECH04725-20
Mean scores of descriptive panels for each attribute of frozen dairy desserts made with 20%, 30%, and 40% sweetpotato puree. A 15-cm (5.9 inches) line scale was used for each attribute: 0 = very weak and 15 = very strong. Tukey’s honestly significant difference test was used to separate treatment means (P ≤ 0.05). Means for each attribute followed by the same letter do not differ.
Citation: HortTechnology hortte 31, 3; 10.21273/HORTTECH04725-20
Analysis of frozen desserts (phase two)
Chemical analysis.
The pH of the frozen dessert samples did not differ among samples and ranged from 6.5 to 6.9 (Table 4). The Brix level of the almond-, milk-, and soy-based frozen desserts were 27.3, 31.2, and 30.1, respectively. The fat content of the samples averaged 10%. The almond milk used in this study contained 0.4% protein, whereas the soy milk contained 2.9% and milk contained 3.8%, thus resulting in the differences in TSS and Brix within desserts. The almond milk-based frozen desserts contained 90 mg/serving of calcium, and the soy- and milk-based frozen dessert contained 68 mg/serving of calcium (Table 5). The potassium contents in the soy-, almond-, and milk-based desserts were 199, 148, and 143 mg/serving, respectively. In all samples, the iron content was not detectable and the sodium content was ≤35 mg/serving. The almond milk sample contained 1 g/serving of crude protein, which was half that in the other samples. The crude carbohydrate contents were estimated by subtracting the crude protein and fat contents from the TSS (range, 21–23 g/serving).
pH, Brix, total fat content, and total soluble solids (TSS) for frozen desserts made with 30% sweetpotato puree by weight and milk, almond milk, or soy milk.
Estimated nutritional content for a 100-g serving of frozen desserts made with 30% sweetpotato puree by weight and milk, almond milk, or soy milk based on the USDA food composition database.z
Despite the nutritional variability among samples, adding sweetpotato puree greatly increased their nutritional value and decreased the added sugar. A 30-g portion of sweetpotato baked in skin contains 288.3 µg of vitamin A retinol activity equivalents (USDA, 2018). A 100-g serving of the frozen dessert samples containing 30 g of sweetpotato puree fulfills ≈50% of the daily recommended intake of vitamin A. The total energy values ranged from 168 to 182 kcal (Table 5).
Consumer acceptability.
Consumer acceptability data from Mississippi State and Pontotoc were evaluated separately. Panelists attending the Sweetpotato Field Day (Pontotoc) moderately liked (7.1) the frozen desserts regardless of the type of base used (Table 6). However, their knowledge of sweetpotatoes may have resulted in bias. These panelists gave higher appearance scores for milk (7.6) than for almond milk (7.1) and soy milk (7.0). The samples were given similar ratings for the other attributes. The panelists at the Mississippi State University Sensory Laboratory indicated that the appearance and texture of the milk-based frozen dessert (7.6 and 7.2, respectively) were preferred over almond-based (6.5 and 6.4, respectively) and soy-based (6.8 and 6.5, respectively) frozen desserts. Regarding aroma, panelists gave the milk-based frozen dessert (6.3) and soy-based frozen dessert (6.1) similar ratings, and they preferred both of those over the almond-based frozen dessert (5.7). The overall acceptability ratings were 6.9 for milk, 6.5 for soy milk, and 6.1 for almond milk.
Consumer acceptability of frozen desserts containing 30% sweetpotato by weight conducted at Mississippi State University (MSU), Mississippi State, MS (n = 101), and Pontotoc, MS (n = 43) in 2018, and analyzed separately by location.
Cluster analysis of consumer acceptability of frozen desserts.
Because of the lack of differences among Pontotoc panelists, a cluster analysis was conducted for the Mississippi State panelists only. Consumer panelists at the Mississippi State University Sensory Laboratory were clustered into four groups based on the overall acceptability of frozen desserts determined by a dendrogram. The mean scores based on clusters (Table 7) were as follows. Cluster 1 comprised 57.4% of the consumer panelists. These panelists did not have a preference among the three samples, with mean overall acceptability ratings of the group ranging between 7.1 and 7.4 and all three desserts rated as moderately liked. Cluster 2 comprised 22.8% of the panelists who provided similar overall acceptability ratings for milk (6.9) and soy milk (6.5). Both were slightly liked and preferred over almond milk (3.9). Cluster 3 comprised 7.9% of panelist who rated all three samples as either slightly disliked or moderately disliked (range, 3.3–4.4) without differences among the samples. Cluster 4 comprised 11.9% of the panelists who preferred milk (7.2) and almond milk (7.4) samples over soy milk (3.8).
Mean scores for overall consumer acceptability (n = 101) of sweetpotato-based frozen dessert samples containing 30% sweetpotato by weight according to different clusters of consumer segments at the Mississippi State University Garrison Sensory Laboratory, Mississippi State, MS, in 2018.
Clusters 1, 2, and 4 rated the milk-based frozen dessert between 6.9 and 7.2; therefore, the treatment was rated as slightly liked or greater by the most panelists (92.1%). The soy milk frozen dessert was scored as 6 or greater by 80.2% of panelists, and the almond milk dessert sample was scored as 6 or greater by 69.3% of panelists. Both milk and soy milk samples were slightly liked or greater by 80.2% of panelists. Milk and almond milk samples were slightly liked or more by 69.3% of panelists. Panelists who liked both the almond and soy samples comprised 57.4% of panelists.
The frozen desserts had similar fat contents but variable protein and sugar contents because of the different compositions of raw materials in the base milk/milk alternative. Results of the present study differ from those of Alozie and Udofia (2015), who reported that consumers preferred almond milk over soy milk in terms of color, flavor, taste, and overall acceptability, but that they gave similar ratings for mouthfeel for the two. However, Alozie and Udofia (2015) investigated the nutritional and sensory properties of vanilla-flavored almond milk and soy milk as beverages, not frozen desserts. In the present study, the lack of corresponding vanilla-flavored frozen desserts formulated without sweetpotato puree and using milk, almond milk, and soy milk did not permit for a potential investigation of the interactive effects of milk/milk alternative base and the presence/absence of sweetpotato puree. Furthermore, given the documented variability in consumer sensory attributes among commercial vanilla-flavored ice creams (Dooley et al., 2010), direct comparisons between the sweetpotato-based frozen desserts in this study and a vanilla ice cream standard were not possible.
During this study, it was determined that frozen desserts could contain as much as 30% sweetpotato puree. When ratings were averaged across all panelists, the sweetpotato-based frozen desserts made from milk and almond milk or soy milk had similar overall acceptability. However, the cluster analysis revealed that preferences among the three formulations differed greatly by cluster. Frozen sweetpotato desserts containing soy milk and almond milk may not be as broadly acceptable as those containing milk. Marketing efforts should be considered for those who are lactose-intolerant, allergic to milk protein, vegan, and consumers who appreciate the health benefits of sweetpotato and coconut oil.
Units
Literature cited
Ahuja, K. & Rawat, A. 2018 Non-dairy ice cream market size by source, by flavor, by product, by form, by distribution channel industry analysis report, regional outlook, growth potential, price trend, competitive market share & forecast, 2018-2024. Global Market Insights, Shelbyville, DE
Alozie, Y.E. & Udofia, U.S. 2015 Nutritional and sensory properties of almond (Prunus amygdalus Var. Dulcis) seed milk World J. Dairy Food Sci. 10 117 121 doi: 10.5829/idosi.wjdfs.2015.10.2.9622
American Oil Chemists’ Society 1998 Official methods and recommended practices of the American Oil Chemists’ Society. 3rd ed. Amer. Oil Chem. Soc., Chicago, IL
Association of Official Agricultural Chemists International 1999 Official methods of analysis of AOAC international. 16th ed. 5th rev. Assn. Offic. Agr. Chem. Intl., Gaithersburg, MD
Bisla, G. , Verma, P. & Sharma, S. 2012 Development of ice creams from soybean milk & watermelon seeds milk and evaluation of their acceptability and nourishing potential Adv. Appl. Sci. Res. 3 371 376
Bonvell-Benjamin, A.C. 2007 Sweet potato: A review of its past, present, and future role in human nutrition Adv. Food Nutr. Res. 52 1 59 doi: 10.1016/S1043-4526(06)52001-7
Burri, B.J. 2011 Evaluating sweet potato as an intervention food to prevent vitamin A deficiency Compr. Rev. Food Sci. Food Saf. 10 118 130 doi: 10.1111/j.1541-4337.2010.00146.x
Choo, S.Y. , Leong, S.K. & Henna Lu, F.S. 2010 Physicochemical and sensory properties of ice-cream formulated with virgin coconut oil Food Sci. Technol. Intl. 16 531 541 doi: 10.1177/1082013210367546
Collart, A.J. , Meyers, S.L. & Ward, J.K. 2019 Consumer perception of skinning injury in sweetpotatoes and implications for marketability: An experimental auction HortTechnology 29 468 475 doi: 10.21273/HORTTECH04355-19
Dooley, L. , Lee, Y.S. & Meullenet, J.F. 2010 The application of check-all-that-apply (CATA) consumer profiling to preference mapping of vanilla ice cream and its comparison to classical external preference mapping Food Qual. Prefer. 21 394 401 doi: 10.1016/j.foodqual.2009.10.002
Goff, H.D. 1997 Colloidal aspects of ice cream - A review Intl. Dairy J. 7 363 373 doi: 10.1016/S0958-6946(97)00040-X
Jenkins, D.J.A. , Wolever, T.M.S. & Jenkins, A.L. 1988 Starchy foods and glycemic index Diabetes Care 11 149 159 doi: 10.2337/diacare.11.2.149
Johnson, T. , Wilson, N. , Worosz, M.R. , Fields, D. & Bond, J.K. 2015 Commodity highlight: Sweet potatoes. U.S. Dept. Agr., Econ. Res. Serv. Pub. No.VGS-355-SA1
King, B.M. 1994 Sensory profiling of vanilla ice cream: Flavour and base interactions Lebensm. Wiss. Technol. 27 450 456 doi: 10.1006/fstl.1994.1091
Lawless, H.T. & Heymann, H. 2010 Sensory evaluation of food: Principles and practices. 2nd ed. Springer-Verlag, Ithaca, NY
Mäkinen, O.E. , Wanhalinna, V. , Zannini, E. & Arendt, E.K. 2016 Foods for special dietary needs: Non-dairy plant-based milk substitutes and fermented dairy-type products Crit. Rev. Food Sci. Nutr. 56 339 349 doi: 10.1080/10408398.2012.761950
Marten, B. , Pfeuffer, M. & Schrezenmeir, J. 2006 Medium-chain triglycerides Intl. Dairy J. 16 1374 1382 doi: 10.1016/j.idairyj.2006.06.015
Meyers, S.L. 2019 State report - Mississippi. Proc. Natl. Sweetpotato Collaborators Group Annu. Mtg. 37:32
Morgan, K.L. , Hood, K. & Myles, A. 2012 Mississippi sweet potato 2012 industry evaluation. Mississippi State Univ. Ext. Serv. Publ. No. 2734
Ohmes, R.L. , Marshall, R.T. & Heymann, H. 1998 Sensory and physical properties of ice creams containing milk fat or fat replacers J. Dairy Sci. 81 1222 1228
Palacios, O.M. , Badran, J. , Drake, M.A. , Reisner, M. & Moskowitz, H.R. 2009 Consumer acceptance of cow’s milk versus soy beverages: Impact of ethnicity, lactose tolerance, and sensory preference segmentation J. Sens. Stud. 24 731 748 doi: 10.1111/j.1745-459X.2009.00236.x
Santana, I.A. , Ribeiro, E.P. & Iguti, A.M. 2011 Evaluation of green coconut (Cocos nucifera L.) pulp for use as milk, fat and emulsifier replacer in ice cream Procedia Food Sci. 1 1447 1453 doi: 10.1016/j.profoo.2011.09.214
Takahata, Y. , Noda, T. & Nagata, T. 1993 HPLC determination of β-carotene content of sweet potato cultivars and its relationship with color values Jpn. J. Breed. 43 421 427 doi: 10.1270/jsbbs1951.43.421
Tufts University 2019 Calculating calories and nutrients in meals. 26 Oct. 2020. <https://hnrca.tufts.edu/flipbook/resources/restaurant-meal-calculator/>
U.S. Department of Agriculture 2018 USDA food composition database. 26 Oct. 2020. <https://ndb.nal.usda.gov/index.html>
U.S. Department of Agriculture-National Agricultural Statistics Service 2018 QuickStats database. 26 Oct. 2020. <https://quickstats.nass.usda.gov/>
U.S. Food and Drug Administration 2018 Requirements for specific standardized frozen desserts (21CFR135). <https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/cfrsearch.cfm?fr=135.110>
