Data scarcity is a major challenge when training deep learning (DL) models. DL demands a large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate data to train DL frameworks. Usually, manual labeling is needed to provide labeled data, which typically involves human annotators with a vast background of knowledge. This annotation process is costly, time-consuming, and error-prone. Usually, every DL framework is fed by a significant amount of labeled data to automatically learn representations. Ultimately, a larger amount of data would generate a better DL model and its performance is also application dependent. This issue is the main barrier for many applications dismissing the use of DL. Having sufficient data is the first step toward any successful and trustworthy DL application. This paper presents a holistic survey on state-of-the-art techniques to deal with training DL models to overcome three challenges including small, imbalanced datasets, and lack of generalization. This survey starts by listing the learning techniques. Next, the types of DL architectures are introduced. After that, state-of-the-art solutions to address the issue of lack of training data are listed, such as Transfer Learning (TL), Self-Supervised Learning (SSL), Generative Adversarial Networks (GANs), Model Architecture (MA), Physics-Informed Neural Network (PINN), and Deep Synthetic Minority Oversampling Technique (DeepSMOTE). Then, these solutions were followed by some related tips about data acquisition needed prior to training purposes, as well as recommendations for ensuring the trustworthiness of the training dataset. The survey ends with a list of applications that suffer from data scarcity, several alternatives are proposed in order to generate more data in each application including Electromagnetic Imaging (EMI), Civil Structural Health Monitoring, Medical imaging, Meteorology, Wireless Communications, Fluid Mechanics, Microelectromechanical system, and Cybersecurity. To the best of the authors’ knowledge, this is the first review that offers a comprehensive overview on strategies to tackle data scarcity in DL.
Background: The purposes of this study were to determine the photogrammetric soft tissue facial profile measurements for Iraqi adults sample with class I normal occlusion using Standardized photographic techniques and to verify the existence of possible gender differences. Materials and methods: Eighty Iraqi adult subjects (40 males and 40 females) with an age ranged between 18-25 years having class I normal occlusion were chosen for this study. Each individual was subjected to clinical examination and digital standardized right side photographic records were taken in the natural head position which is mirror position which the patient looking straight into his eyes into the mirror mounted on the stand. The photographs were analyzed using A
... Show MoreThe effects of using aqueous nanofluids containing covalently functionalized graphene nanoplatelets with triethanolamine (TEA-GNPs) as novel working fluids on the thermal performance of a flat-plate solar collector (FPSC) have been investigated. Water-based nanofluids with weight concentrations of 0.025%, 0.05%, 0.075%, and 0.1% of TEA-GNPs with specific surface areas of 300, 500, and 750 m2/g were prepared. An experimental setup was designed and built and a simulation program using MATLAB was developed. Experimental tests were performed using inlet fluid temperatures of 30, 40, and 50 °C; flow rates of 0.6, 1.0, and 1.4 kg/min; and heat flux intensities of 600, 800, and 1000 W/m2. The FPSC’s efficiency increased as the flow rate and hea
... Show MorePituitary adenomas are the anterior pituitary tumors. Patients with an Aryl Hydrocarbon Receptor-Interacting Protein (AIP) mutation (AIP- mut) tend to have more aggressive tumors occurring at a younger age. Single nucleotide polymorphisms (SNPs) in many studies have been related to metabolic comorbidities in the general population. Study aims investigated the role of AIP gene SNPs with susceptibility to acromegaly pituitary- adenoma, with levels of LH, FSH, TSH, Testosterone, IGF1,GH, FT4 , Prolactin hormones and blood sugar levels. The study was conducted on a group of acromegaly patients, including 50 patients) both Genders( with hyperplasia of the ends, and apparently healthy control group. Genotyping of
... Show More
Background: Diabetes mellitus is a chronic disease with an increasing prevalence worldwide and characterized by an increase in oxidative stress and inflammation. The most important factor that is responsible for oxidative stress and production of reactive oxygen species (ROS) is hyperglycemia. The major targets of ROS are proteins. The most common and widely used biomarker of severe oxidative protein damage is protein carbonyl content.
The study was designed to assess the serum level of protein carbonyl as a marker of protein oxidation in patients with type 2 diabetes mellitus and to evaluate the effect of age, body weight, waist circumference, diabetic control and disease duration on the level
... Show MoreThis study numerically intends to evaluate the effects of arc-shaped fins on the melting capability of a triplex-tube confinement system filled with phase-change materials (PCMs). In contrast to situations with no fins, where PCM exhibits relatively poor heat response, in this study, the thermal performance is modified using novel arc-shaped fins with various circular angles and orientations compared with traditional rectangular fins. Several inline and staggered layouts are also assessed to maximize the fin’s efficacy. The effect of the nearby natural convection is further investigated by adding a fin to the bottom of the heat-storage domain. Additionally, the Reynolds number and temperature of the heat-transfer fluid (HTF) are e
... Show MoreThis study concerns the role of activated carbon (AC) from palm raceme as a support material for the enhancement of lipase-catalyzed reactions in an aqueous solution, with deep eutectic solvent (DES) as a co-solvent. The effects of carbonization temperature, impregnation ratio, and carbonization time on lipase activity were studied. The activities of Amano lipase from Burkholderia cepacia (AML) and lipase from the porcine pancreas (PPL) were used to investigate the optimum conditions for AC preparation. The results showed that AC has more interaction with PPL and effectively provides greater enzymatic activity compared with AML. The optimum treatment conditions of AC samples that yield the highest enzymatic activity were 0.5 (NaOH (
... Show MoreIn this paper, a novel coronavirus (COVID-19) model is proposed and investigated. In fact, the pandemic spread through a close contact between infected people and other people but sometimes the infected people could show two cases; the first is symptomatic and the other is asymptomatic (carrier) as the source of the risk. The outbreak of Covid-19 virus is described by a mathematical model dividing the population into four classes. The first class represents the susceptible people who are unaware of the disease. The second class refers to the susceptible people who are aware of the epidemic by media coverage. The third class is the carrier individuals (asymptomatic) and the fourth class represents the infected ind
... Show MoreThis research studies the effect of addition of some nanoparticles
(MgO, CuO) and grain size (30,40nm) on some physical properties
(impact strength, hardness and thermal conductivity) for a matrix
blend of epoxy resin with SBR rubber. Hand –Lay up method was
used to prepare the samples. All samples were immersed in water for
9 weeks.
The Results showed decreased in the values of impact strength and
hardness but increased the coefficient of thermal conductivity.