Wellness rediscovered: a complete approach towards a healthy life Essay

People are unaware of the calories that they intake from their everyday meals. Keeping track of these calories is an essential step towards a healthy life. With Wellness Rediscovered, we envisage helping the user for maintaining a Nutrition based e-diary for physical well-being. Wellness Rediscovered focuses on telling the user how close or far they are from their goal. With the conceptual fundamentals of Image Processing and Machine Learning, we propose to provide rudimentary analysis of the food eaten, by simply analyzing the photo taken and measure the amount of nutrients, calories, fats intake automatically. The proposed system will identify the food item from the image and calculate the calories for same by making use of a special calibration technique. Users can take pictures of their meals every day and the system will log these calories in a database. By using a regression algorithm on the logged data, a detailed analysis report is generated.

Introduction

A calorie is a measuring unit which is defined as the amount of heat energy needed to raise the temperature of one gram of water by one degree. The process of providing or obtaining the food necessary for health and growth is called Nutrition. This unit is commonly used to measure the overall amount of energy in any food portion that consists of the main food components of Carbohydrate, Protein and Fat. Calories are a must for the body, as they generate energy. But it is said that an excess of anything is bad and the same applies to the intake of calories too. If there is an excess of calories in our body, it gets stored in the form of fats, thus making us overweight. Adult calorie requirements differ from that of a child and in the same way, the daily calorie requirement of a Body Mass Index is a person’s weight in kilograms divided by the square of their height in meters. It is one of the most commonly used ways of estimating whether a person is overweight or not. A person is considered obese when his / her BMI is higher than or equal to 30 kg/m2. The rate of obese person is increasing in an alarming rate from the last few years. Also there are many chances for obese people to face a serious health problems like hypertension, heart attack, diabetes, obesity, hypertension, high cholesterol etc. So the main cause for obesity is imbalance of the amount of food intake and energy consumed by the individual since it is necessary to have healthy meal. Therefore, different systems were developed which would measure the nutrition level of the diet and helps the patients and dietitians to control their obesity. This system reviews the different systems which had taken the food images to measure the calorie and nutritional level in the food sample. As such, this system is use to measure the amount of calories consumed in a meal would be of great help not only to patients and dietitians in the treatment of obesity, but also to the calorie conscious person. Obesity treatment needs the patient to note the amount of the daily food intake, but in most cases, it is not simple for the patients to measure or control their daily intake due to the lack of nutrition, education or self control. Therefore, by using a automatic food intake monitoring system, we can assist the patient and provide an effective tool for the obesity treatment. Nowadays, new technologies such as computers and smart phones are involved in the medical treatment of different types of diseases, and obesity is considered as one of the common disease. From the last few years, a numbers of food intake measuring methods have been developed. But most of these systems have drawbacks such as large calculation errors and it is not an user friendly. In this project, there is no need of large calculation for calorie and nutrition measurement.Obesity in adults has become a serious problem. A person is considered obese when the Body Mass Index is higher than or equal to 30 (kg/m2 ) [1]. In 2008, more than one in ten of the world’s adult populations were obese [1], but in 2012 this figure has risen to one in six adults [2], an alarming growth rate. Recent studies have shown that obese people are more likely to have serious health conditions such as hypertension, heart attack, type II diabetes, high cholesterol, breast and colon cancer, and breathing disorders. The main cause of obesity is the imbalance between the amount of food intake and energy consumed by the individuals [3]. So, in order to lose weight in a healthy way, as well as to maintain a healthy weight for normal people, the daily food intake must be measured [4]. In fact, all existing obesity treatment techniques require the patient to record all food intakes per day in order to compare the food intake to consumed energy. But, in most cases, unfortunately patients face difficulties in estimating and measuring the amount of food intake due to the self-denial of the problem, lack of nutritional information, the manual process of writing down this information (which is tiresome and can be forgotten), and other reasons. As such, a semi-automatic monitoring system to record and measure the amount of calories consumed in a meal would be of great help not only to patients and dietitians in the treatment of obesity, but also to the average calorie-conscious person. Indeed, a number of food intake measuring methods have been developed in the last few years. But, most of these systems have drawbacks such as usage difficulties or large calculation errors. Furthermore, many of these methods are for experimental practices and not for real life usage, as we shall see in the section II. In this paper, we propose a personal software instrument to measure calorie and nutrient intake using a smartphone or any other mobile device equipped with a camera. Our system uses image processing and segmentation to identify food portions (i.e., isolating portions such as chicken, rice, vegetables, etc., from the overall food image), measures the volume of each food portion, and calculates nutritional facts of each portion by calculating the mass of each portion from its measured volume and matching it against existing nutritional fact tables. While a preliminary description of our work has been presented in [5], here we extend it by proposing a more accurate measurement method for estimating food portion volume, which also works for food portions with an irregular shape, and by evaluating our approach with more food items. More importantly, the segmentation features are enriched by involving texture as well as color, shape and size of the objects. Our results show reasonable accuracy in the estimation of nutritional values of food types for which our system has been trained. Color and texture are fundamental characters of natural images, and play an important role in visual perception. Color has been used in identifying objects for many years. Texture is one of the most active topics in machine intelligence and pattern analysis since the 1950s which tries to discriminate different patterns of images by extracting the dependency of intensity between pixels and their neighboring pixels [6], or by obtaining the variance of intensity across pixels [7]. Recently, different features of color and texture are combined together in order to measure food nutrition more accurately [8]. In our proposed system, we also aim at using smartphones as monitoring tools as they are widely accessible and easy to use.

Related Work

In tradition, we can track our meals by taking note or using some applications on mobile devices to record our meal. This method is simple and suitable for everyone. However, users must rely on their memory to record what they had in each meal. Although the users might record immediately after their meal, they might record the wrong information. For example, they might know that eating 2-3 pieces of pizza can make them fatter, so they change the meal information to avoid suffer from the real information. To avoid the problems discussed before, many researchers have been trying to solve these problems by creating an algorithm that can deliver the amount of calories in automatic or semi-automatic way. Given the widespread use of mobile devices such as digital cameras and smartphones, these devices can now be considered as data collection tools for dietitians [1]. With the benefit of image processing techniques, some researchers proposed vision-based approach to identify the amount of calories taken by the users. There are some existing works that proposed the method which aimed at photo-based food recording. [2] presented a system for detecting food images and estimating the food balance by categorizing food into grains, vegetables, meatlfish/beans, fruit, and dairy products. However, this method could not be used with Thai-food image because, normally, Thai food is not served by separating food materials as Japanese food or Western food. [3] proposed calorie-content estimation based on low-level image features, whereby a food photo is visually searched using the low level. In this paper, we developed based on this approach. In this study, we tried to find some suitable features for Thai-food images recognition. Sensor-based techniques been also investigated. [4] presented a novel approach that uses an electroglottograph (EGG) device for an objective and automatic detection of food intake. Variations in the electrical impedance across the larynx caused by the passage of food during swallowing were captured by the EGG device. Artificial intelligence techniques are used to recognize the type of food that have been swallowed. Although this method could deliver more accuracy result, the users are required to use some specific devices with high cost and too far from their real life.A daily diet is very necessary in day to day life. So it is necessary to manage our daily food item intake. In 2008 to 2010, more than one in ten of the world’s adult populations were obese [1], but in 2012 this figure or range has risen to one in six adults [2], an alarming growth rate. The recent paper studies have shown that obese people are more likely to have serious health conditions such as hypertension, heart attack, diabetes, high cholesterol, breast and colon cancer, and breathing disorders, thyroid etc. The main cause of obesity is the imbalance between the amount of daily food intake and energy consumed by the individuals [3]. There is another system which is based on support vector machine but use the thumb for calibration of each and every food image but it require long calculation for measuring nutrition that measurement system also uses a photo of the food, taken with the camera of a smart phone, but uses the thumb of patient for calibration, which solves the problem of carrying cards or special trays. More specifically, an thumb image is captured and stored with its measurements in the first usage time (first time calibration). Now, this unique method will lead to relatively accurate results without the difficulties of other methods. Food images will then be taken with the user’s thumb placed next to the dish, make it easy to measure the real life size of the portions. We then apply image processing and classification techniques to find the food portions, their volume and area of the food and get the calorie and nutrition but the use of thumb is necessary[4]. So, in order to lose weight in a healthy way, as well as to maintain a healthy weight for normal people, the daily food intake measured is must [5]. That system's are uses image processing and segmentation to identify food portions (i.e., isolating portions such as chicken, rice, vegetables, etc., from the overall food image), measures the volume of each food part, and calculates nutritional facts of each part by calculating the mass of each portion from its measured volume[6]. Color is used in identifying objects for many years and also Texture is one of the most active topics in machine intelligence and pattern analysis since the 1950s which tries to discriminate the different patterns of images by extracting the dependency of intensity between pixels and their neighboring pixels [7], or by obtaining the variance of intensity across pixels [8]. Recently, different features of color ",texture, size are combined together in order to measure food nutrition more accurately [9]. The problem with this manual approach is obvious people not remembering exactly what they are ate, forgetting to take note, and needing to see an expert dietician on a very frequent basis so the dietician can guess how much calories and nutrient the patient has taken. To evaluate the shortcomings of these clinical methods, researchers have been trying to come up with new improved techniques. Some of these techniques require the person to take a picture of the food before eating food, so that the picture can be processed offline, either manually or automatically, to measure the amount of calorie. For example, the work in [10] proposes a method that uses a calibration card for an reference, this card should be placed next to the food when capturing the image, so that the dimensions of the food are known. However, this card must always be present in the photo when the patient or obese people wants to use the system. The drawback is of the system will not work without this card, which means that in the case of absence of the card, the system will not work. Another method which use the photo of the food and feeds that to a Neural Network developed by researchers in [11]. But the user must have to capture the photo in a special tray (for calibration purpose), which might not be always possible and so the method may be difficult to follow for the user. That system also uses an photo of the food, taken with the camera of a smart phones, but uses the patient’s thumb for calibration, which solves the problem of carrying cards or special trays. More specifically, an image of the thumb is captured and stored with its measurements in the first time usage (first time calibration). Different food images will then be taken with the user’s thumb placed next to the dish, makes it easy to measure the real size of the portions. We then apply image processing and classification techniques to find the food portions, their volume, and their nutritional facts. Yet another approach appears in [12], where the picture of the food taken with a smart phone is compared to photos of predefined foods with known nutritional values which are stored in a database, and the values are estimated based on picture similarity. The main disadvantage of this system is that it does not take into account the size and shape of the food, which is extremely important. One example, which is typical of current clinical approaches, is the 24-Hour Dietary Recall system. The idea of this method is the listing of the daily food intake by using a special format for a period of 24 hours. This method requires a trained interviewer, such as a dietician, to ask the respondent to remember in details all the food and drinks s/he has consumed during a period of time in the recent past or a whole day (often the previous 24 hours). The 24HR requires only short-term memory, and if the recall is unannounced, the diet is not changed for that person. Also, the interview is relatively brief (20 to 30 minutes), and the subject burden is as less in comparison with other food image recording methods [13].

Proposed System

  • · User interacts with the mobile app and captures an image.
  • · The image is passed to the server application, which uses the Watson Visual Recognition Service to analyze the images and Nutritionix API to provide nutritional information.
  • · Data is returned to the Android mobile app for display.
  • · The information of food intake with quantity is tracked.
  • · The user can set target weight gain / weight loss
  • · Our system also predicts amount of calories that one would need to consume each day in order to achieve the target weight.
  • · The system also provides graphical representation of calorie intake day wise.

Conclusion And Future Work

In this project, we proposed a measurement method that estimates the amount of calories from a food’s image by measuring the volume of the food portions from the image and using nutritional facts tables to measure the amount of calories and nutrition in the food. As we argued, our system is designed to aid dieticians for the treatment of obese or overweight people, although normal people can also benefit from our system by controlling more closely their daily eating without worrying about overeating and weight gain. We focused on identifying food items in an image by using IBM Watson API, food portion volume measurement, and calorie measurement based on food portion mass and nutritional tables. Our results indicated reasonable accuracy of our method in area measurement, and subsequently volume and calorie measurement. An obvious avenue for future work is to cover more food types from a variety of cuisines around the world. Also, more work is needed for supporting mixed or even liquid food, if possible.

How to cite this essay: