An official method for measuring GI was issued by the International Organization for Standardization (ISO) in 2010 [22] after development and update [21,23]. The ISO method has been calibrated by three independent laboratories, Sydney University, Australia, and GI labs in Toronto, Canada, and Biofortis Merieux NutriSciences in Saint-Herblain, France. Using the 2010 ISO method is precise enough to differentiate a low-GI food from a high-GI food with a high probability (97–99%) [24].

The in vivo method of GI measurement is summarized in Table 1. Participant recruitment [24], test sample preparation [22], blood sample collection [22], and data analysis [25,26] are key steps for this ISO method.

Prior to in vivo tests for GI measurement, test foods are usually analyzed to meet stringent nutritional criteria for energy (kJ or kcal), carbohydrate, saturated fats, sodium, and (in certain foods) fibre and calcium [7,8,9].

The in vivo method of GI measurement has high accuracy and precision [23,24]. At the same time, it requires extra effort to manage the participants and obtain ethics clearance [22]. The cost of the in vivo test is relatively high [10], especially for food formulae, which are still under development. Other researchers have explored alternative methods to determine the GI values of foods, mainly in vitro methods. The GI value determined by an in vitro method is sometimes incorrect when classifying a low-GI food, and to label a high-GI food as a low-GI category is potentially harmful for people with diabetes [18]. So far, the GI measurement of single foods is still determined by in vivo methods [18].

Besides the glycemic effect, the in vitro method also covers the understanding of food nutrition and formulas, food digestibility and other health benefits by mimicking food digestion in living bodies. Many studies have focused on the chemical analysis of the food digestion process [20,27,28,29]. The chemical analysis, either for physical models or for mathematical models [30,31,32,33,34], has not always considered shear stresses and shear rates when mimicking the digestion process [35]. A comprehensive review of all digestion models is covered by a related paper [35] (Sections 2 and 4 in [35]). Recent digestion models focus on mimicking the kinetics of food movement in the digestive system, as well as the physical processes during the peristaltic movement of digestion system. It is worthwhile to understand food digestion (especially carbohydrate digestion) from engineering perspectives to pave the way for an improved in vitro methods of GI measurement.

Starch is the majority carbohydrate in plants and is deposited in granules in most green plants. Its hydrolysis provides 40–80% of the total human energy intake [21,36]. It is found in many types of plant tissues and organs, such as seeds (e.g., cereal grains), roots (e.g., sweet potato), tubers (e.g., potato), stems (e.g., sago), leaves (e.g., tobacco), fruits (e.g., banana), and even pollen [37,38]. Starch is the dominant component of cereal grains, pluses, and tuber and root crops [37]. For instance, milled rice kernels contain up to 90% starch on a dry basis [39], maize kernels contain up to 80% starch [40], and potatoes contain 60–80% starch [41]. Besides starch-rich crops, pulse grains, such as legumes, have up to 53% starch [42]. Starch is a biopolymer. It contains two major components: amylose and amylopectin. Amylose is a mainly linear polysaccharide, which contributes up to 15–35% of the granules. Amylopectin, however, is a highly branched polysaccharide [37,38,43]. Amylose, containing α-1,4-linked d-glucopyranose and a few branches of α-1,6 linkages, has different properties to amylopectin with α-1,4-linked linear chains of different lengths, connected by about 5% α-1,6 branch linkages [38,44,45,46]. Amylose tends to produce tough gels and strong films, while amylopectin produces soft gels and weak films. It has been reported that a high amylose content in starch may help reduce the glycemic response and increase the blood glucose level slowly [47].

Recent studies have advanced the understanding of the starch features and the fine structure of amylose and amylopectin [37,48]. The importance of plant cell walls and the cellulose tissue structure have been noticed by several works [49,50,51], and these features have also been discussed in the studies of carbohydrate digestion [52,53,54]. The cell wall may be an important mass-transfer resistance from a mass transfer point of view. Starch granules transfer through the broken cell wall before its hydrolysis with digestive enzymes during the digestion processes. This section may be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn.

Studying human digestion was initiated in the medical field for diagnostic purposes [55]. Nowadays, it is an essential factor in the development of novel food products, as well as the testing of new pharmaceutical products [56,57].

The process of food digestion generally includes oral digestion, esophageal transit, gastric digestion, small intestinal digestion, and large intestinal fermentation. Oral digestion is the initial process to produce a bolus, which is a mass of chewed food [58,59]. The bolus has small particle sizes for safe swallowing [60,61]. Once a bolus is swallowed, it moves by esophageal peristalsis, as well as by the simple force of gravity when not lying down [62]. Then, the bolus passes into the stomach. The stomach plays the role as a food container, mixer, grinder, and sieve [63]. The majority of food breakdown happens in the stomach, where the bolus blends with gastric acid as well as digestive enzymes. The speed of food breakdown in the stomach is important to determine other digestion processes, such as gastric emptying, as well as nutrient absorption [57]. Food moves into the small intestine after it moves out of the stomach. Further food breakdown happens in the small intestine. The partially digested food from the stomach is broken down into small molecules to be absorbed and carried into the bloodstream. The large intestine is colonized by microorganisms, which ferment food particles that have not been digested completely. Only water and fermentation by-products are absorbed in the large intestine [62].

Knowledge about the human digestion process may benefit from research in the medical and nutrition fields. However, the fundamental mechanisms are still not completely understood. For instance, Glycemic Index measurement is common in nutritional studies and is conducted by a physiological method [24,64]. The factors related to the glycemic response still require quantitative understanding in terms of the physical and chemical properties of foods. Mass transfer theory may help provide the quantitative analysis by understanding the processes of mass transfer and chemical reaction during food digestion processes. Such engineering perspectives may contribute to the fundamental understanding of food digestion.

The digestion process of starch can be partially quantified by the rate of starch loss, the rate of glucose appearance, and the rate of appearance of various oligosaccharides [65]. It has been stated that “Understanding the factors influencing starch digestion is best done through a causal, mechanistically based approach through the following paradigm: biosynthesis → growth and processing conditions → structure of starch and of starch-containing substances → digestion properties” [65].

In the human digestive system, starch is catalyzed by salivary amylase and pancreatic amylase, which are both α-amylases (Enzyme Commission number is 3.2.1.1) [66]. Salivary amylase is the first enzyme for starch hydrolysis in the mouth [57,67]. This process occurs over a relatively short time (within one minute). When the bolus of food moves into the stomach, the action of α-amylase slows down and the acid hydrolysis of starch increases. The hydrolysis of starch in the stomach may also be affected by the residual activity of salivary amylase, and the acidity of the stomach is likely to partly reduce the activity of the salivary amylase [68]. From the stomach to the duodenum, the bolus encounters α-amylase the pancreatic secretion, which contains sodium hydrogen carbonate and α-amylase. Sodium hydrogen carbonate neutralizes the acidic fluid from the stomach to a pH of about 8 [69]. The continues the catalysis of starch into disaccharides and oligosaccharides. The oligosaccharides, such as α-limit dextrins, small linear oligomers, and larger α-glucans are not absorbed into the blood stream until their further hydrolysis to glucose. In the small intestine, enzymes, including mucosal maltase–glucoamylase and sucrase–isomaltase, catalyze the oligosaccharides into single glucose [65,68].

After a meal, the peak plasma glucose response usually occurs within the first hour, and the glucose level increase seldom lasts more than two hours. This observation puts a clear emphasis on the mouth, salivary fluids, and the stomach as features that may be very important in the glycemic response. At the same time, the pancreas secretes insulin and inhibits the release of glucagon, so that the glucose is normally taken up by muscle and fat tissue. Plasma glucose levels have a range between 3.3 and 8.3 mmol/L, providing body energy for the organs and tissues. However, high postprandial glucose levels are related to the development of Type 2 diabetes and/or cardiovascular disease in susceptible persons [3,70,71]. People with diabetes have a high blood glucose level (hyperglycemia) due to deficiencies in insulin secretion or in insulin action [15,71]. The current in vivo method to determine the glycemic response of a meal/food is to measure small numbers of blood samples from the finger over a period of two hours. The Glycemic Index is then calculated to classify carbohydrate-rich foods [24]. Understanding starch digestion and absorption of starch-derived glucose may help in the maintenance of stable plasma glucose levels.