scholarly journals An Intelligent Mobile Application to Automate the Analysis of Food Calorie using Artificial Intelligence and Deep Learning

2021 ◽  
Author(s):  
Yongqing Yu ◽  
Yishan Zou ◽  
Yu Sun
Keyword(s):  
Eating Habits ◽  
Learning Rate ◽  
Classification Model ◽  
Batch Size ◽  
Calorie Intake ◽  
Dynamic Learning ◽  
Standard Size ◽  
Control And Prevention ◽  
Food Image

As obesity becomes increasingly common worldwide [9], more and more people want to lose weight – for both their health and their image. According to the Centers for Disease Control and Prevention (CDC), long-term changes in daily eating habits (such as regarding food/nutrition type, calorie intake) are successful at keeping weights off [10]. Therefore, it would be helpful to have an AI mobile program that identifies the types of food the user consumes and automatically calculates the total calories. This paper examines the development and optimization of an 11-categorical food classification model based on the MobileNet neural network using Python. Specifically, it classifies any food image as one of bread, dairy, dessert, egg product, fried food, meat, noodles, rice, seafood, soup, or fruit/vegetables. Methods of optimization include data preprocessing and learning rate and batch size adjustments. Experimental results show that scaling image inputs to standard size (Python Numpy resize() function), 300 training epochs, dynamic learning rate (start with 0.001 and *0.1 for every 30 epochs), and a batch size of 16 yields our best model of 83.44% accuracy.

scholarly journals An 11-Categorical AI Food Classification Model Based on Mobile-Net Neural Networks

2021 ◽  
Vol 3 (4) ◽  
pp. 93-103
Author(s):  
Yongqing Yu ◽  
Yishan Zou ◽  
Yu Sun
Keyword(s):  
Eating Habits ◽  
Learning Rate ◽  
Classification Model ◽  
Batch Size ◽  
Calorie Intake ◽  
Dynamic Learning ◽  
Standard Size ◽  
Model Based ◽  
Control And Prevention ◽  
Food Image

As obesity becomes increasingly common worldwide [1], more people want to lose weight to improve their health and image. According to the Centers for Disease Control and Prevention (CDC), long-term changes in daily eating habits (such as regarding food/ nutrition type, calorie intake) are successful at keeping weights off [2]. Therefore, it would be helpful to have an artificial intelligence (AI) mobile program that identifies the types of food the user consumes and automatically calculates the total calories. This paper examines the development and optimization of an 11-categorical food classification model based on the Mobile-Net neural network using Python. Specifically, it classifies any food image as one of bread, dairy, dessert, egg product, fried food, meat, noodles, rice, seafood, soup, or fruit/vegetables. Methods of optimization include data preprocessing and learning rate and batch size adjustments. Experimental results show that scaling image inputs to standard size (Python Numpy resize) function), 300 training epochs, dynamic learning rate (start with 0.001 and *0.1 for every 30 epochs), and a batch size of 16 yields our best model of 83.44% accuracy.

scholarly journals Treating Traumatic Brain Injuries with Electroceuticals: Implications for the Neuroanatomy of Consciousness

NeuroSci ◽  
2021 ◽  
Vol 2 (3) ◽  
pp. 254-265
Author(s):  
Jihad Aburas ◽  
Areej Aziz ◽  
Maryam Butt ◽  
Angela Leschinsky ◽  
Marsha L. Pierce
Keyword(s):  
Brain Stimulation ◽  
Brain Injuries ◽  
Vegetative State ◽  
Minimally Conscious State ◽  
Conscious State ◽  
Permanent Disability ◽  
United States Food ◽  
Control And Prevention ◽  
Minimally Conscious

According to the Centers for Disease Control and Prevention (CDC), traumatic brain injury (TBI) is the leading cause of loss of consciousness, long-term disability, and death in children and young adults (age 1 to 44). Currently, there are no United States Food and Drug Administration (FDA) approved pharmacological treatments for post-TBI regeneration and recovery, particularly related to permanent disability and level of consciousness. In some cases, long-term disorders of consciousness (DoC) exist, including the vegetative state/unresponsive wakefulness syndrome (VS/UWS) characterized by the exhibition of reflexive behaviors only or a minimally conscious state (MCS) with few purposeful movements and reflexive behaviors. Electroceuticals, including non-invasive brain stimulation (NIBS), vagus nerve stimulation (VNS), and deep brain stimulation (DBS) have proved efficacious in some patients with TBI and DoC. In this review, we examine how electroceuticals have improved our understanding of the neuroanatomy of consciousness. However, the level of improvements in general arousal or basic bodily and visual pursuit that constitute clinically meaningful recovery on the Coma Recovery Scale-Revised (CRS-R) remain undefined. Nevertheless, these advancements demonstrate the importance of the vagal nerve, thalamus, reticular activating system, and cortico-striatal-thalamic-cortical loop in the process of consciousness recovery.

Neuroendocrine regulation of energy balance: Implications on the development and surgical treatment of obesity

2017 ◽  
Vol 23 (3) ◽  
pp. 131-146 ◽  
Author(s):  
Gisele Farias ◽  
Bárbara Dal Molin Netto ◽  
Solange Cravo Bettini ◽  
Ana Raimunda Dâmaso ◽  
Alexandre Coutinho Teixeira de Freitas
Keyword(s):  
Weight Loss ◽  
Adipose Tissue ◽  
Gastrointestinal Tract ◽  
Energy Balance ◽  
Public Health Problem ◽  
Neuroendocrine Regulation ◽  
Calorie Intake ◽  
Hormone Signaling ◽  
Sustained Weight Loss

Introduction: Obesity, a serious public health problem, occurs mainly when food consumption exceeds energy expenditure. Therefore, energy balance depends on the regulation of the hunger–satiety mechanism, which involves interconnection of the central nervous system and peripheral signals from the adipose tissue, pancreas and gastrointestinal tract, generating responses in short-term food intake and long-term energy balance. Increased body fat alters the gut- and adipose-tissue-derived hormone signaling, which promotes modifications in appetite-regulating hormones, decreasing satiety and increasing hunger senses. With the failure of conventional weight loss interventions (dietary treatment, exercise, drugs and lifestyle modifications), bariatric surgeries are well-accepted tools for the treatment of severe obesity, with long-term and sustained weight loss. Bariatric surgeries may cause weight loss due to restriction/malabsorption of nutrients from the anatomical alteration of the gastrointestinal tract that decreases energy intake, but also by other physiological factors associated with better results of the surgical procedure. Objective: This review discusses the neuroendocrine regulation of energy balance, with description of the predominant hormones and peptides involved in the control of energy balance in obesity and all currently available bariatric surgeries. Conclusions: According to the findings of our review, bariatric surgeries promote effective and sustained weight loss not only by reducing calorie intake, but also by precipitating changes in appetite control, satiation and satiety, and physiological changes in gut-, neuro- and adipose-tissue-derived hormone signaling.

scholarly journals Online Model Updating and Dynamic Learning Rate-Based Robust Object Tracking

Sensors ◽  
2018 ◽  
Vol 18 (7) ◽  
pp. 2046 ◽  
Author(s):  
Md Islam ◽  
Guoqing Hu ◽  
Qianbo Liu
Keyword(s):  
Object Tracking ◽  
Model Updating ◽  
Learning Rate ◽  
Dynamic Learning

scholarly journals FETAL TYPE CLASSIFICATION BASED ON FETAL ELECTROCARDIOGRAPH IN NON-STRESS TESTS

2021 ◽  
pp. 2140052
Author(s):  
JIANLI LIU ◽  
YIMIN YANG ◽  
SONG ZHANG ◽  
XUWEN LI ◽  
LIN YANG ◽  
...  
Keyword(s):  
Heart Rate ◽  
Fetal Heart Rate ◽  
Fetal Heart ◽  
Classification Model ◽  
Support Vector ◽  
Stress Tests ◽  
Nonlinear Parameters ◽  
Design Algorithm ◽  
Long Time

Electronic fetal heart rate (FHR) monitoring is a technical means to evaluate the state of the fetus in the uterus by monitoring FHR. The main purpose is to detect intrauterine hypoxia and take corresponding medical measures timely. Because the fetus sleeps quietly for up to 1 hour sometimes, ultrasound Doppler is not easy to continuously detect for a long time. The electronic fetal monitor obtains the fetal heart rate, which not only improves the accuracy and comfort, but also the convenient implementation of long-term monitoring. It is beneficial to reduce perinatal fetal morbidity and mortality. This study used maternal–fetal Holter monitor which is based on the technology of fetal electrocardiograph (FECG) to collect the FHR, and then design algorithm to extract the baseline FHR, acceleration, variation, sleep-wake cycle and nonlinear parameters. There were significant differences in the 22 parameters between the normal and the suspicious group. Using the 22 characteristic parameters, the support vector machine was used to classify the normal and the suspected group of fetuses. 80% of the data was used to train a classification model. The remaining 20% of the data was used as a test set and its accuracy reached 93.75%.

scholarly journals A léböjtkúrák és az atípusos evészavarok lehetséges kapcsolata. Kvalitatív előtanulmány

2018 ◽  
Vol 159 (28) ◽  
pp. 1153-1157 ◽  
Author(s):  
Enikő Bóna ◽  
Attila Forgács ◽  
Ferenc Túry
Keyword(s):  
Eating Behaviors ◽  
Healthy Lifestyle ◽  
Eating Habits ◽  
Holistic Approach ◽  
The Body ◽  
Orthorexia Nervosa ◽  
Calorie Intake ◽  
Rapid Weight Loss ◽  
Clinical Environment ◽  
Purging Disorder

Abstract: Introduction and aim: There are two notable eating behaviors that are not far from having their own category as a mental disorder: the purging disorder, that is already among the DSM-5 non-specific eating disorders, and orthorexia nervosa, when a person is fixating too much on healthy foods. Our purpose is to describe how these can be observed in recreational juice cleanse camps, which are very popular today as an alternative health trend. Method: The first author recorded her data during multisited ethnographic observations in two Hungarian juice cleanse camps. Based on the diary logs, notes and interviews collected, we will present the motives of eating anomalies that the participants had shown. Results: The main motive of the camp is “detoxification”. The lack of solid food, drastically low calorie intake and lots of physical activity will bring an inevitable change in the body, that is interpreted as toxins leaving the body. Participants have also included deliberate use of laxatives in their everyday routines, with which they associate positive connotations and are linked to the spiritual processes of “letting it go” and “renewal” in the spirit of a holistic approach. The use of symbols in the physiological processes was highly noticeable. Rapid weight loss due to diuresis, the desire for “clean” meals, and “self-rewarding” borrowed from the esoteric-self-help culture are also common motives. Due to the refeeding complications, so far two deaths have been reported by camp organizers. Conclusions: Both purging disorder and orthorexia nervosa can be well-identified in our observations. This shows that also in the non-clinical environment, there is an institutionalization of eating habits that are dangerous to the health. This “detox” is not only physiologically harmful, but it is not proved to provide long-term help in mental health either. As a solution, we advocate developing an appropriate health communication plan for misconceptions about healthy lifestyle and eating, and also a promotion of psychotherapeutic opportunities. Orv Hetil. 2018; 159(28): 1153–1157.

Uncertainty Quantification and Optimization of Deep Learning for Fracture Recognition

2021 ◽  
Author(s):  
Ryan Santoso ◽  
Xupeng He ◽  
Marwa Alsinan ◽  
Hyung Kwak ◽  
Hussein Hoteit
Keyword(s):  
Deep Learning ◽  
Uncertainty Quantification ◽  
Learning Rate ◽  
Batch Size ◽  
Global Maximum ◽  
Fractured Reservoir ◽  
Training Set ◽  
Fast Learning ◽  
Learning Rates ◽  
Aleatoric Uncertainty

Abstract Automatic fracture recognition from borehole images or outcrops is applicable for the construction of fractured reservoir models. Deep learning for fracture recognition is subject to uncertainty due to sparse and imbalanced training set, and random initialization. We present a new workflow to optimize a deep learning model under uncertainty using U-Net. We consider both epistemic and aleatoric uncertainty of the model. We propose a U-Net architecture by inserting dropout layer after every "weighting" layer. We vary the dropout probability to investigate its impact on the uncertainty response. We build the training set and assign uniform distribution for each training parameter, such as the number of epochs, batch size, and learning rate. We then perform uncertainty quantification by running the model multiple times for each realization, where we capture the aleatoric response. In this approach, which is based on Monte Carlo Dropout, the variance map and F1-scores are utilized to evaluate the need to craft additional augmentations or stop the process. This work demonstrates the existence of uncertainty within the deep learning caused by sparse and imbalanced training sets. This issue leads to unstable predictions. The overall responses are accommodated in the form of aleatoric uncertainty. Our workflow utilizes the uncertainty response (variance map) as a measure to craft additional augmentations in the training set. High variance in certain features denotes the need to add new augmented images containing the features, either through affine transformation (rotation, translation, and scaling) or utilizing similar images. The augmentation improves the accuracy of the prediction, reduces the variance prediction, and stabilizes the output. Architecture, number of epochs, batch size, and learning rate are optimized under a fixed-uncertain training set. We perform the optimization by searching the global maximum of accuracy after running multiple realizations. Besides the quality of the training set, the learning rate is the heavy-hitter in the optimization process. The selected learning rate controls the diffusion of information in the model. Under the imbalanced condition, fast learning rates cause the model to miss the main features. The other challenge in fracture recognition on a real outcrop is to optimally pick the parental images to generate the initial training set. We suggest picking images from multiple sides of the outcrop, which shows significant variations of the features. This technique is needed to avoid long iteration within the workflow. We introduce a new approach to address the uncertainties associated with the training process and with the physical problem. The proposed approach is general in concept and can be applied to various deep-learning problems in geoscience.

scholarly journals The Relation between Eating Habits and Abdominal Fat, Anthropometry, PON1 and IL-6 Levels in Patients with Multiple Sclerosis

Nutrients ◽  
2020 ◽  
Vol 12 (3) ◽  
pp. 744 ◽  
Author(s):  
Eraci Drehmer ◽  
Jose Luis Platero ◽  
Sandra Carrera-Juliá ◽  
Mari Luz Moreno ◽  
Asta Tvarijonaviciute ◽  
...  
Keyword(s):  
Multiple Sclerosis ◽  
Abdominal Obesity ◽  
Eating Habits ◽  
Paraoxonase 1 ◽  
Calorie Intake ◽  
Fat Percentage ◽  
Food Diary ◽  
Carbohydrate Consumption ◽  
Caloric Density ◽  
Elisa Technique

Background: Multiple sclerosis (MS) is a chronic neurodegenerative disease of an inflammatory, demyelinating and autoimmune nature. Diets with a high caloric density could be especially relevant in terms of the pathogenesis related to an increase in adipose tissue that is metabolically active and releases mediators, which can induce systemic inflammation and an increased oxidation state. The aim of this study was to analyse the eating habits related to calorie intake and their impact on abdominal obesity associated with anthropometric variables, the activity of the oxidation marker paraoxonase 1 (PON1), and interleukin 6 (IL-6) levelsin MS patients. Methods: An analytical and quantitative observational study was conducted with a population of 57 MS patients. The dietary-nutritional anamnesis was gained through the Food Frequency Questionnaire and a food diary. Diet and eating habits have been analysed through the Easy Diet–Programa de gestión de la consulta® software. Anthropometric measurements were taken in order to determine the presence of abdominal obesity. In addition, PON1 was quantified in serum by means of automated spectrophotometric assays and IL-6 was quantified using the ELISA technique. Results: A normal calorie intake was determined for women, yet a slightly lower intake was observed in men. Carbohydrate consumption was below what was established, and protein and lipids were over, in both cases. Furthermore, most patients had abdominal obesity, with significantly higher body mass index (BMI), waist-to-hip ratio (WHR), waist-to-height ratio (WHtR), fat percentage and IL-6 levels. IL-6 is greatly correlated with waist circumference and WHtR. Conclusion: MS patients’ nutrient intake shows an imbalance between macronutrients. This seems to favour the abdominal obesity associated with high values of proinflammatory IL-6 that is not correlated with a lower activity of PON1.

Sign in / Sign up

Export Citation Format

Share Document