A Survey of Imitation Learning: Algorithms, Recent Developments, and Challenges.

In recent years, the development of robotics and artificial intelligence (AI) systems has been nothing short of remarkable. As these systems continue to evolve, they are being utilized in increasingly complex and unstructured environments, such as autonomous driving, aerial robotics, and natural language processing. As a consequence, programming their behaviors manually or defining their behavior through the reward functions as done in reinforcement learning (RL) has become exceedingly difficult. This is because such environments require a high degree of flexibility and adaptability, making it challenging to specify an optimal set of rules or reward signals that can account for all the possible situations. In such environments, learning from an expert's behavior through imitation is often more appealing. This is where imitation learning (IL) comes into play -a process where desired behavior is learned by imitating an expert's behavior, which is provided through demonstrations.This article aims to provide an introduction to IL and an overview of its underlying assumptions and approaches. It also offers a detailed description of recent advances and emerging areas of research in the field. Additionally, this article discusses how researchers have addressed common challenges associated with IL and provides potential directions for future research. Overall, the goal of this article is to provide a comprehensive guide to the growing field of IL in robotics and AI.

Robust and scalable single-molecule protein sequencing with fluorosequencing.

The need to accurately survey proteins and their modifications with ever higher sensitivities, particularly in clinical settings with limited samples, is spurring development of new single molecule proteomics technologies. Fluorosequencing is one such highly parallelized single molecule peptide sequencing platform, based on determining the sequence positions of select amino acid types within peptides to enable their identification and quantification from a reference database. Here, we describe substantial improvements to fluorosequencing, including identifying fluorophores compatible with the sequencing chemistry, mitigating dye-dye interactions through the use of extended polyproline linkers, and developing an end-to-end workflow for sample preparation and sequencing. We demonstrate by fluorosequencing peptides in mixtures and identifying a target neoantigen from a database of decoy MHC peptides, highlighting the potential of the technology for high sensitivity clinical applications.

An Uncertainty-Aware Transfer Learning-Based Framework for COVID-19 Diagnosis.

The early and reliable detection of COVID-19 infected patients is essential to prevent and limit its outbreak. The PCR tests for COVID-19 detection are not available in many countries, and also, there are genuine concerns about their reliability and performance. Motivated by these shortcomings, this article proposes a deep uncertainty-aware transfer learning framework for COVID-19 detection using medical images. Four popular convolutional neural networks (CNNs), including VGG16, ResNet50, DenseNet121, and InceptionResNetV2, are first applied to extract deep features from chest X-ray and computed tomography (CT) images. Extracted features are then processed by different machine learning and statistical modeling techniques to identify COVID-19 cases. We also calculate and report the epistemic uncertainty of classification results to identify regions where the trained models are not confident about their decisions (out of distribution problem). Comprehensive simulation results for X-ray and CT image data sets indicate that linear support vector machine and neural network models achieve the best results as measured by accuracy, sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve (AUC). Also, it is found that predictive uncertainty estimates are much higher for CT images compared to X-ray images.

Robust Adaptive Control Scheme for Teleoperation Systems With Delay and Uncertainties.

This paper proposes a robust adaptive algorithm that effectively copes with time-varying delay and uncertainties in Internet-based teleoperation systems. Time-delay induced by the communication network, as a major problem in teleoperation systems, along with uncertainties in modeling of robotic manipulators and remote environment warn the stability and performance of the system. A robust adaptive control algorithm is developed to deal with the system uncertainties and to provide a smooth estimation of delayed reference signals. The proposed control algorithm generates chattering-free torques which is one of the practical considerations for robotic applications. In addition, the achieved input-to-state stability gains do not necessarily require high gain control torques to retain the system's stability. Experimental simulation studies validate the effectiveness of the proposed control strategy on a teleoperation system consisting of a Phantom Omni Haptic device and SimMechanics model of the industrial manipulator UR10. The validation of the proposed control methodology was executed through a real-time Internet-based communication established over 4G mobile networks between Australia and Scotland.

Machine learning-based coronary artery disease diagnosis: A comprehensive review.

Coronary artery disease (CAD) is the most common cardiovascular disease (CVD) and often leads to a heart attack. It annually causes millions of deaths and billions of dollars in financial losses worldwide. Angiography, which is invasive and risky, is the standard procedure for diagnosing CAD. Alternatively, machine learning (ML) techniques have been widely used in the literature as fast, affordable, and noninvasive approaches for CAD detection. The results that have been published on ML-based CAD diagnosis differ substantially in terms of the analyzed datasets, sample sizes, features, location of data collection, performance metrics, and applied ML techniques. Due to these fundamental differences, achievements in the literature cannot be generalized. This paper conducts a comprehensive and multifaceted review of all relevant studies that were published between 1992 and 2019 for ML-based CAD diagnosis. The impacts of various factors, such as dataset characteristics (geographical location, sample size, features, and the stenosis of each coronary artery) and applied ML techniques (feature selection, performance metrics, and method) are investigated in detail. Finally, the important challenges and shortcomings of ML-based CAD diagnosis are discussed.

Biochemical and Structural Insights into the Preference of Nairoviral DeISGylases for Interferon-Stimulated Gene Product 15 Originating from Certain Species.

UNLABELLED: The regulation of the interferon type I (IFN-I) response has been shown to rely on posttranslational modification by ubiquitin (Ub) and Ub-like interferon-stimulated gene product 15 (ISG15) to stabilize, or activate, a variety of IFN-I signaling and downstream effector proteins. Unlike Ub, which is almost perfectly conserved among eukaryotes, ISG15 is highly divergent, even among mammals. Since zoonotic viruses rely on viral proteins to recognize, or cleave, ISG15 conjugates in order to evade, or suppress, innate immunity, the impact of ISG15 biodiversity on deISGylating proteases of the ovarian tumor family (vOTU) from nairoviruses was evaluated. The enzymatic activities of vOTUs originating from the Crimean-Congo hemorrhagic fever virus, Erve virus, and Nairobi sheep disease virus were tested against ISG15s from humans, mice, shrews, sheep, bats, and camels, which are mammalian species known to be infected by nairoviruses. This along with investigation of binding by isothermal titration calorimetry illustrated significant differences in the abilities of nairovirus deISGylases to accommodate certain species of ISG15. To investigate the molecular underpinnings of species preferences of these vOTUs, a structure was determined to 2.5 Å for a complex of Erve virus vOTU protease and a mouse ISG15 domain. This structure revealed the molecular basis of Erve virus vOTU's preference for ISG15 over Ub and the first structural insight into a nonhuman ISG15. This structure also revealed key interactions, or lack thereof, surrounding three amino acids that may drive a viral deISgylase to prefer an ISG15 from one species over that of another. IMPORTANCE: Viral ovarian tumor domain proteases (vOTUs) are one of the two principal classes of viral proteases observed to reverse posttranslational modification of host proteins by ubiquitin and interferon-stimulated gene product 15 (ISG15), subsequently facilitating downregulation of IFN-I signaling pathways. Unlike the case with ubiquitin, the amino acid sequences of ISG15s from various species are notably divergent. We illustrate that vOTUs have clear preferences for ISG15s from certain species. In addition, these observations are related to the molecular insights acquired via the first X-ray structure of the vOTU from the Erve nairovirus in complex with the first structurally resolved nonhuman ISG15. This information implicates certain amino acids that drive the preference of vOTUs for ISG15s from certain species.

Prevalence of palpable thyroid nodule in Isfahan, Iran, 2006: a population based study.

SUBJECTS: This cross-sectional study enrolled 2 523 of adult general population (age > 20 years) of Isfahan city. History taking and thyroid exam was performed by seven trained general practitioners. The concentration of TSH was measured in all, TgAb and TPOAb in approximately one in six and urinary iodine concentration in one fourth of them, randomly. We recalled people with any suspicious abnormality in thyroid exam or whose TSH levels were not normal at present or past. They were re-visited by an endocrinologist to confirm or rule out the diagnosis and second serum sample was obtained to measure T4, T3, T3RU and TSH. RESULTS: Thyroid nodule was identified in 62 (17 men and 45 women) out of 2 523 participants (2.5%; 95% CI: 2-3). Single and multiple nodules were found in 56 (2.2%; 95% CI: 1-2.5) and 6 (0.24%; 95% CI: 0.07-0.5) out of 2 523 persons, respectively. The prevalence of nodule was higher in females (3.5%) than in males (1.3%) (OR=2.72, 95% CI, 1.53-5.06, P=0.001). It was significantly higher in patients older than 35 years in comparison with younger people (3.3% vs. 1.3%; OR=2.5; 95% CI, 1.3-4.7; P=0.002). CONCLUSION: The prevalence of thyroid nodule in our study is less than what has been reported in other countries. It is due to lower prevalence in female population. Its prevalence in males is similar to other studies. It seems that geographical and racial differences or higher prevalence of deeper location of some nodules in female Isfahani population could explain this finding.

Contrasting effects of calcium channel blockade versus converting enzyme inhibition on proteinuria in African Americans with non-insulin-dependent diabetes mellitus and nephropathy.

Hypertension is a common finding in non-insulin-dependent diabetes mellitus (NIDDM) nephropathy. African Americans have a high prevalence of NIDDM and hypertension, and are relatively resistant to the antihypertensive effects of converting enzyme inhibitors (CEI) but respond well to calcium channel blockers (CCB). In the long-term study presented here, the effects of isradipine, a dihydropyridine calcium antagonist, on the course of the nephropathy were investigated and compared with the effects of captopril in 31 African Americans with NIDDM and proteinuria (> or = 500 mg/day). The patients were stratified by levels of GFR and proteinuria, and they were randomized to receive isradipine (N = 16) or captopril (N = 15); doses were adjusted to maintain similar BP levels (< 140/90). At 6 months, mean arterial pressure was similar (102 +/- 3 and 104 +/- 3 mm Hg in the isradipine and captopril groups, respectively) and GFR was unchanged (delta = -4 +/- 3 and +1 +/- 3 ml/min/1.73 in the isradipine and captopril groups, respectively; P = NS). However, proteinuria in the isradipine group increased by approximately 50% (2.01 +/- 0.40 versus 3.04 +/- 0.70 mg/mg creatinine at baseline versus 6 months, respectively, P < 0.05), whereas captopril reduced proteinuria by 30% after 6 months (2.85 +/- 0.70 at baseline versus 2.30 +/- 0.70 mg/mg creatinine, P < 0.05). Dietary protein, sodium intake, and HbA1C levels were similar in both groups and did not differ from baseline. It was concluded that over 6 months, captopril reduces and isradipine increases proteinuria in African Americans with NIDDM and nephropathy. Whether this contrasting effect on proteinuria will result in different rates of progression is not known, but dihydropyridine CCB should be used cautiously in African Americans with diabetic nephropathy.

Optical measurements of lung microvascular filtration coefficient using polysulfone fibers.

Lung fluid balance, which is governed by the product of net transvascular pressure difference and lung filtration coefficient, can be altered in pulmonary diseases. A simple measurement of the lung filtration coefficient (Kfc) would be clinically useful and has been examined by several researchers. Current methods of determining Kfc include gravimetric measurement in isolated lungs and lymph node cannulation, neither of which can be extended to human use. Optical measurements of protein concentration changes in venous blood can be combined with pressure measurements to calculate Kfc. Blood, though, contains red corpuscles, which tend to absorb and scatter light, obscuring these optical measurements. In this study, an optical system was developed in which a polysulfone filter cartridge was used to remove red blood cells before the filtrate was passed through a spectrophotometer. Absorbance changes caused by changes in concentration of albumin labeled with Evans Blue were monitored at 620 nm after venous pressure was elevated by about 13 cm H2O. Optical measurements of Kfc averaged 0.401 +/- 0.074 (ml/min cm H2O 100 g DLW) for an isolated canine lung. Optical measurements of Kfc (0.363 +/- 0.120 ml/min cm H2O 100 g DLW) were made for the first time in an intact, closed chest sheep in which pulmonary pressure was altered by inflating a Foley balloon in the left atrium. We conclude that absorbance and scattering artifacts introduced by red blood cells can be eliminated by first filtering the blood through polysulfone fibers. Kfc measurements using the optical method are similar to values obtained by others using gravimetric methods. Finally, we have demonstrated that the technique can be used to estimate Kfc in an intact animal.

Preliminary study of the protective effect of the calcium channel blocker, nifedipine, on adriamycin-induced tissue injury.

The effect of nifedipine, an oral calcium-channel blocker, on adriamycin-induced wound healing was studied. Nifedipine was administered to animals prior to treating with adriamycin. The healing strengths of cecal, fascial, and skin anatomoses were measured. Animals treated with adriamycin alone had significant decreases in healing strength compared to controls: for cecum at postoperative days (POD) 7 and 14; for skin at PODs 14, 21, and 28; for fascia at POD 28. Overall, nifedipine reduced the impact of adriamycin injury on the strength of healing incisional anastomoses. Nifedipine also appeared to protect collagen synthesis at anastomotic sites from adriamycin-induced impairment.

A new microporous membrane for diagnostic immunoassays.

A family of new microporous composite membranes that do not require post-immobilization blocking or quenching steps has been developed for diagnostic immunoassay applications. These membranes, Autobloc membranes, are shown to have low nonspecific binding of proteins. Immobilization of antibody or antigen to the membrane can be achieved either by covalent binding or by adsorption. The correlation between the initial immobilization capacity for IgG and the effective immobilization capacity of microporous membranes is discussed. The performance of Autobloc membranes is demonstrated by dot-enzyme-linked immunoassays in the flow-through mode with several commonly used enzyme-substrate systems and compared to the performance of other membranes. The advantages of using Autobloc membranes are suggested.