Ensemble Wavelet Decomposition-Based Detection of Mental States Using Electroencephalography Signals.

Technological advancements in healthcare, production, automobile, and aviation industries have shifted working styles from manual to automatic. This automation requires smart, intellectual, and safe machinery to develop an accurate and efficient brain-computer interface (BCI) system. However, developing such BCI systems requires effective processing and analysis of human physiology. Electroencephalography (EEG) is one such technique that provides a low-cost, portable, non-invasive, and safe solution for BCI systems. However, the non-stationary and nonlinear nature of EEG signals makes it difficult for experts to perform accurate subjective analyses. Hence, there is an urgent need for the development of automatic mental state detection. This paper presents the classification of three mental states using an ensemble of the tunable Q wavelet transform, the multilevel discrete wavelet transform, and the flexible analytic wavelet transform. Various features are extracted from the subbands of EEG signals during focused, unfocused, and drowsy states. Separate and fused features from ensemble decomposition are classified using an optimized ensemble classifier. Our analysis shows that the fusion of features results in a dimensionality reduction. The proposed model obtained the highest accuracies of 92.45% and 97.8% with ten-fold cross-validation and the iterative majority voting technique. The proposed method is suitable for real-time mental state detection to improve BCI systems.

Advances in Smart Environment Monitoring Systems Using IoT and Sensors.

Air quality, water pollution, and radiation pollution are major factors that pose genuine challenges in the environment. Suitable monitoring is necessary so that the world can achieve sustainable growth, by maintaining a healthy society. In recent years, the environment monitoring has turned into a smart environment monitoring (SEM) system, with the advances in the internet of things (IoT) and the development of modern sensors. Under this scenario, the present manuscript aims to accomplish a critical review of noteworthy contributions and research studies on SEM, that involve monitoring of air quality, water quality, radiation pollution, and agriculture systems. The review is divided on the basis of the purposes where SEM methods are applied, and then each purpose is further analyzed in terms of the sensors used, machine learning techniques involved, and classification methods used. The detailed analysis follows the extensive review which has suggested major recommendations and impacts of SEM research on the basis of discussion results and research trends analyzed. The authors have critically studied how the advances in sensor technology, IoT and machine learning methods make environment monitoring a truly smart monitoring system. Finally, the framework of robust methods of machine learning; denoising methods and development of suitable standards for wireless sensor networks (WSNs), has been suggested.

Deep Learning Neural Network Prediction System Enhanced with Best Window Size in Sliding Window Algorithm for Predicting Domestic Power Consumption in a Residential Building.

Buildings are considered to be one of the world's largest consumers of energy. The productive utilization of energy will spare the accessible energy assets for the following ages. In this paper, we analyze and predict the domestic electric power consumption of a single residential building, implementing deep learning approach (LSTM and CNN). In these models, a novel feature is proposed, the "best N window size" that will focus on identifying the reliable time period in the past data, which yields an optimal prediction model for domestic energy consumption known as deep learning recurrent neural network prediction system with improved sliding window algorithm. The proposed prediction system is tuned to achieve high accuracy based on various hyperparameters. This work performs a comparative study of different variations of the deep learning model and records the best Root Mean Square Error value compared to other learning models for the benchmark energy consumption dataset.

Machine learning-based predictive models for identifying high active compounds against HIV-1 integrase.

HIV-integrase is an important drug target because it catalyzes chromosomal integration of proviral DNA towards establishing latent infection. Computer-aided drug design has immensely contributed to identifying and developing novel antiviral drugs. We have developed various machine learning-based predictive models for identifying high activity compounds against HIV-integrase. Multiclass models were built using support vector machine with reasonable accuracy on the test and evaluation sets. The developed models were evaluated by rigorous validation approaches and the best features were selected by Boruta method. As compared to the model developed from all descriptors set, a slight improvement was observed among the selected descriptors. Validated models were further used for virtual screening of potential compounds from ChemBridge library. Of the six high active compounds predicted from selected models, compounds 9103124, 6642917 and 9082952 showed the most reasonable binding-affinity and stable-interaction with HIV-integrase active-site residues Asp64, Glu152 and Asn155. This was in agreement with previous reports on the essentiality of these residues against a wide range of inhibitors. We therefore highlight the rigorosity of validated classification models for accurate prediction and ranking of high active lead drugs against HIV-integrase.

Ftl-CoV19: A Transfer Learning Approach to Detect COVID-19.

COVID-19 is an infectious and contagious disease caused by the new coronavirus. The total number of cases is over 19 million and continues to grow. A common symptom noticed among COVID-19 patients is lung infection that results in breathlessness, and the lack of essential resources such as testing, oxygen, and ventilators enhances its severity. Chest X-ray can be used to design and develop a COVID-19 detection mechanism for a quicker diagnosis using AI and machine learning techniques. Due to this silver lining, various new COVID-19 detection techniques and prediction models have been introduced in recent times based on chest radiography images. However, due to a high level of unpredictability and the absence of essential data, standard models have showcased low efficiency and also suffer from overheads and complexities. This paper proposes a model fine tuning transfer learning-coronavirus 19 (Ftl-CoV19) for COVID-19 detection through chest X-rays, which embraces the ideas of transfer learning in pretrained VGG16 model with including combination of convolution, max pooling, and dense layer at different stages of model. Ftl-CoV19 reported promising experimental results; it observed training and validation accuracy of 98.82% and 99.27% with precision of 100%, recall of 98%, and F1 score of 99%. These results outperformed other conventional state of arts such as CNN, ResNet50, InceptionV3, and Xception.

Yoga Pose Estimation and Feedback Generation Using Deep Learning.

Yoga is a 5000-year-old practice developed in ancient India by the Indus-Sarasvati civilization. The word yoga means deep association and union of mind with the body. It is used to keep both mind and body in equilibration in all flip-flops of life by means of asana, meditation, and several other techniques. Nowadays, yoga has gained worldwide attention due to increased stress levels in the modern lifestyle, and there are numerous methods or resources for learning yoga. Yoga can be practiced in yoga centers, through personal tutors, and can also be learned on one's own with the help of the Internet, books, recorded clips, etc. In fast-paced lifestyles, many people prefer self-learning because the abovementioned resources might not be available all the time. But in self-learning, one may not find an incorrect pose. Incorrect posture can be harmful to one's health, resulting in acute pain and long-term chronic concerns. In this paper, deep learning-based techniques are developed to detect incorrect yoga posture. With this method, the users can select the desired pose for practice and can upload recorded videos of their yoga practice pose. The user pose is sent to train models that output the abnormal angles detected between the actual pose and the user pose. With these outputs, the system advises the user to improve the pose by specifying where the yoga pose is going wrong. The proposed method was compared to several state-of-the-art methods, and it achieved outstanding accuracy of 0.9958 while requiring less computational complexity.

Investigating Feature Ranking Methods for Sub-Band and Relative Power Features in Motor Imagery Task Classification.

Interpreting the brain commands is now easier using brain-computer interface (BCI) technologies. Motor imagery (MI) signal detection is one of the BCI applications, where the movements of the hand and feet can be recognized via brain commands that can be further used to handle emergency situations. Design of BCI techniques encountered challenges of BCI illiteracy, poor signal to noise ratio, intersubject variability, complexity, and performance. The automated models designed for emergency should have lesser complexity and higher performance. To deal with the challenges related to the complexity performance tradeoff, the frequency features of brain signal are utilized in this study. Feature matrix is created from the power of brain frequencies, and newly proposed relative power features are used. Analysis of the relative power of alpha sub-band to beta, gamma, and theta sub-band has been done. These proposed relative features are evaluated with the help of different classifiers. For motor imagery classification, the proposed approach resulted in a maximum accuracy of 93.51% compared to other existing approaches. To check the significance of newly added features, feature ranking approaches, namely, mutual information, chi-square, and correlation, are used. The ranking of features shows that the relative power features are significant for MI task classification. The chi-square provides the best tradeoff between accuracy and feature space. We found that the addition of relative power features improves the overall performance. The proposed models could also provide quick response having reduced complexity.

Optimizing Residual Networks and VGG for Classification of EEG Signals: Identifying Ideal Channels for Emotion Recognition.

Emotion is a crucial aspect of human health, and emotion recognition systems serve important roles in the development of neurofeedback applications. Most of the emotion recognition methods proposed in previous research take predefined EEG features as input to the classification algorithms. This paper investigates the less studied method of using plain EEG signals as the classifier input, with the residual networks (ResNet) as the classifier of interest. ResNet having excelled in the automated hierarchical feature extraction in raw data domains with vast number of samples (e.g., image processing) is potentially promising in the future as the amount of publicly available EEG databases has been increasing. Architecture of the original ResNet designed for image processing is restructured for optimal performance on EEG signals. The arrangement of convolutional kernel dimension is demonstrated to largely affect the model's performance on EEG signal processing. The study is conducted on the Shanghai Jiao Tong University Emotion EEG Dataset (SEED), with our proposed ResNet18 architecture achieving 93.42% accuracy on the 3-class emotion classification, compared to the original ResNet18 at 87.06% accuracy. Our proposed ResNet18 architecture has also achieved a model parameter reduction of 52.22% from the original ResNet18. We have also compared the importance of different subsets of EEG channels from a total of 62 channels for emotion recognition. The channels placed near the anterior pole of the temporal lobes appeared to be most emotionally relevant. This agrees with the location of emotion-processing brain structures like the insular cortex and amygdala.

An Ensemble Deep Learning Model for Automatic Modulation Classification in 5G and Beyond IoT Networks.

With rapid advancement in artificial intelligence (AI) and machine learning (ML), automatic modulation classification (AMC) using deep learning (DL) techniques has become very popular. This is even more relevant for Internet of things (IoT)-assisted wireless systems. This paper presents a lightweight, ensemble model with convolution, long short term memory (LSTM), and gated recurrent unit (GRU) layers. The proposed model is termed as deep recurrent convoluted network with additional gated layer (DRCaG). It has been tested on a dataset derived from the RadioML2016(b) and comprises of 8 different modulation types named as BPSK, QPSK, 8-PSK, 16-QAM, 4-PAM, CPFSK, GFSK, and WBFM. The performance of the proposed model has been presented through extensive simulation in terms of training loss, accuracy, and confusion matrix with variable signal to noise ratio (SNR) ranging from -20 dB to +20 dB and it demonstrates the superiority of DRCaG vis-a-vis existing ones.

Multi- class classification of breast cancer abnormalities using Deep Convolutional Neural Network (CNN).

The real cause of breast cancer is very challenging to determine and therefore early detection of the disease is necessary for reducing the death rate due to risks of breast cancer. Early detection of cancer boosts increasing the survival chance up to 8%. Primarily, breast images emanating from mammograms, X-Rays or MRI are analyzed by radiologists to detect abnormalities. However, even experienced radiologists face problems in identifying features like micro-calcifications, lumps and masses, leading to high false positive and high false negative. Recent advancement in image processing and deep learning create some hopes in devising more enhanced applications that can be used for the early detection of breast cancer. In this work, we have developed a Deep Convolutional Neural Network (CNN) to segment and classify the various types of breast abnormalities, such as calcifications, masses, asymmetry and carcinomas, unlike existing research work, which mainly classified the cancer into benign and malignant, leading to improved disease management. Firstly, a transfer learning was carried out on our dataset using the pre-trained model ResNet50. Along similar lines, we have developed an enhanced deep learning model, in which learning rate is considered as one of the most important attributes while training the neural network. The learning rate is set adaptively in our proposed model based on changes in error curves during the learning process involved. The proposed deep learning model has achieved a performance of 88% in the classification of these four types of breast cancer abnormalities such as, masses, calcifications, carcinomas and asymmetry mammograms.

Serum vascular endothelial growth factor in breast cancer.

Many studies have reported elevated serum concentrations of vascular endothelial growth factor (VEGF) in patients with cancer and claimed that the measurement of circulating VEGF is a surrogate marker of angiogenesis and/or metastasis. To determine the value of VEGF measurement in the diagnosis and prognosis of breast cancer, we measured levels in women with and without breast cancer. Platelet-depleted plasma VEGF levels were measured in premenopausal women at four-day intervals across the menstrual cycle, postmenopausal women and postmenopausal women who had undergone hysterectomy. Platelet-depleted plasma VEGF was also measured in pre- and postmenopausal women with early breast cancer (EBC) and levels compared with intratumoral levels, clinicopathological prognostic parameters and microvessel density. Levels of VEGF were determined using ELISA and immuno-histochemistry. Microvessel density was determined by immunohistochemical CD34 staining. Plasma VEGF in premenopausal women remained stable across the menstrual cycle except for a peak between days 8 and 12. VEGF levels in postmenopausal women were higher than in premenopausal women unless postmenopausal women had undergone hysterectomy. Amongst premenopausal women, levels of VEGF were high in 22 EBC patients when compared to normal premenopausal controls. No correlation was found between plasma and intratumoral VEGF, clinicopathological prognostic parameters or tumour microvessel density. The origin of circulating VEGF differs between pre- and postmenopausal women. Its measurement is unlikely to provide clinically useful diagnostic and prognostic information in women with early and advanced breast cancer.

p53 and chromosome 3 abnormalities, characteristic of malignant lung tumours, are detectable in preinvasive lesions of the bronchus.

Bronchial epithelial dysplasia is believed to precede invasive squamous cell carcinoma of the lung. Six paired dysplasia and tumour samples were distinguished histologically in sections of formalin-fixed paraffin-embedded lung tissue from patients with lung cancer. Additionally, samples of dysplastic bronchial epithelium were obtained from patients without lung tumours. Microdissection of the unstained sections provided dysplastic and tumour samples from which DNA was prepared for comparison with the patients' constitutional genotype, using polymerase chain reaction-based restriction fragment length polymorphism analysis. All six samples of tumour and the paired adjacent samples of bronchial dysplasia showed loss of heterozygosity (LOH) at loci on the short arm of chromosome 3. Five of the six cases showed involvement of the p53 gene as assessed by LOH at the AccII site within the gene, and by immunoreactivity to CM-1, an antibody which recognizes the mutated form of the p53 protein in paraffin-embedded material. Of the dysplastic samples, obtained from patients without invasive tumours, all three showed LOH at 3p; one sample showed LOH at the AccII polymorphic locus within the p53 gene, and another sample, uninformative at this locus, stained positively with this antibody. These results indicate that somatic genetic changes are present in preinvasive lesions in the bronchus.

Injectable midazolam as oral premedicant in pediatric neurosurgery.

In a randomized, double blind, placebo controlled study; the acceptability, efficacy and safety of injectable midazolam as oral premedicant in children was evaluated. One hundred children (ASA 1,2) aged 6 months to 6 years, undergoing elective neurosurgical operations, like meningomyelocele, meningo-encephalocele, ventriculo peritoneal and other shunts and craniotomies for tumour decompression etc., were included in the study. The patients were randomly assigned to one of four groups (A, B, C, D) receiving respectively saline or 0.50, 0.75 and 1.0 mg/kg midazolam in honey, 45 min before separation from parents. All received identical general anesthesia (GA). Age, sex, weight, heart rate, blood pressure, respiratory rate, saturation (SaO2), reaction to parent's separation, sedation score and duration of anesthesia, recovery conditions and side effects were noted. We found no difference in age, sex, weight, patient acceptability vomiting after ingestion and duration of anesthesia between groups. Even though many children resisted the placement of premedicant in the mouth, only three children spat it out and none vomited after swallowing. The reaction to separation from parents was better after midazolam premedication. However, on reaching the operating room, 24% children (placebo-60%) were found anxious after 0.50 mg/kg, but 12% were deeply sedated after a dose of 1.0 mg/kg. Recovery was similar in groups A, B and C except that more (48%) patients were anxious in group A. Recovery, however was delayed in 16% patients of group D. Though, fewer complications were reported during recovery after midazolam than placebo premedication, they were minimal in the 0.75 mg/kg group. We concluded that giving injectable midazolam orally as premedication in pediatric age group scheduled for neurosurgical operations is acceptable, effective and safe in 0.75 mg/kg dose. While 0.50 mg/kg is less effective, 1.0 mg/kg does not offer any additional benefit over 0.75 mg/kg but does delay recovery and may compromise safety.

Influence of nanosized confinements on 4-n-decyl-4'-cyanobiphenyl (10CB): a broadband dielectric study.

Real (epsilon') and imaginary (epsilon") parts of the complex dielectric permittivity (epsilon*) of the liquid crystal (LC) 4-n-decyl-4'-cyanobiphenyl (10CB) embedded in Anopore membranes and Vycor porous glass, as well as dispersed with hydrophilic aerosils, have been studied by means of broadband dielectric spectroscopy in the frequency range from 10(-2) Hz to 1 GHz . In bulk 10CB, which has a direct transition from an isotropic to a smectic- A phase, there exists one main relaxation process for the parallel orientation of the director with respect to the probing field and a faster one for the perpendicular orientation. All molecular relaxation processes in 10CB are of Debye type and have Arrhenius like temperature dependence. For 10CB embedded in untreated and surface treated cylindrical pores of Anopore membranes the dielectric spectra are similar to the bulk with the exception that both the rotation around the short axis and the libration motion are faster in the pores. In the case of 10CB dispersed with two different concentrations of hydrophilic aerosils an emergence of a slow relaxation process, which is stronger for the higher concentration, is additionally observed along with the bulklike processes. The slow process in the LC-hydrophilic aerosil system is attributed to the relaxation of the molecules that are homeotropically aligned close to the surfaces of the aerosil particles. This process also has an Arrhenius type of temperature dependence. For 10CB embedded in narrow channels of Vycor porous glass three relaxation processes are observed. Two of these processes are bulklike and are due to the librational motion of molecules and the rotation of molecules around their short axes. The slowest process seems to be a surface process, similar in origin to that observed for 10CB dispersed with hydrophilic aerosils, and is prominent amongst all. The material in the Vycor porous glass could be supercooled by at least 185 degrees below bulk crystallization temperature. The slow process has a Vogel-Fulcher-Tamman (VFT) type of temperature dependence typical for glass formers in this wide temperature range. In addition, the bulklike processes are found to be strongly modified and also have a VFT like temperature dependence from measurements done in a wide temperature range. This behavior is in contrast to other reports of glassy behavior in confined LC, where the glassy behavior as concluded from a slow relaxation process observed in a narrow temperature range.

Dielectric spectroscopy of aerosil-dispersed liquid crystal embedded in Anopore membranes.

The complex dielectric permittivity epsilon* values are presented for aerosil-dispersed 4-pentyl-4-cyanobiphenyl (5CB) confined in Anopore membranes. The dielectric permittivities are measured in the frequency range from 10(-2) Hz to 1 GHz at temperatures ranging from 50 degrees C down to -20 degrees C. In bulk 5CB, which has only a nematic phase, there exist two main relaxation processes: one due to the rotation of molecules around their short axes for parallel orientation of the director to the probing field and another fast relaxation process due to the librational motion of molecules for perpendicular orientation. Inside Anopore membranes both these main relaxation processes can be observed, but with subtle differences. The relaxation process due to the rotation of molecules around the short axis is faster in Anopores at all temperatures in comparison with the bulk process. Hydrophilic aerosil particles, when dispersed in the liquid-crystal (LC) phase, attach to each other via hydrogen bonds and form a three-dimensional interconnecting aerosil network, thus dividing the LC phase into small domains. Dispersing 5CB with different concentrations of hydrophilic aerosils leads to a decrease in relaxation time with aerosil concentration. In these dispersed systems a slow additional relaxation process emerges. This slow process becomes stronger with higher concentrations of aerosil. From our experiments we conclude that this process is the relaxation of 5CB molecules homeotropically aligned to the surface of the aerosil particles. In the case of 5CB-aerosil dispersions enclosed in Anopore membranes this slow process still exists and increases also with aerosil concentration. The relaxation time of the rotation of the 5CB molecules around their short axis systematically increases in these 5CB-aerosil samples in Anopore membranes with aerosil concentration from the 5CB-Anopore behavior towards the behavior observed for 5CB-aerosil dispersions. We explain the evolution as resulting from opposing tendencies from size effects (in the Anopore membranes) and disorder effects (in the aerosil dispersions).

Shedding of oocysts of Cryptosporidium in immunocompetent patients.

Forty nine patients (19 adults and 30 children) with oocysts of Cryptosporidium in their faeces had repeated stool specimens taken until oocysts could no longer be identified. They were found in the stools up to 35 days after the onset of symptoms in one patient, but most had stopped shedding them by 20 days. In 25 of the 49 patients in whom symptoms could be compared with the shedding of oocysts, 19 (76%) had symptoms corresponding to the shedding period while symptoms persisted in four (16%) after shedding had stopped.

Two-hour versus six-hour administration of amphotericin B in the treatment of kala-azar.

In a randomised study, 50 parasitologically confirmed cases of kala-azar were allocated to two treatment groups: group A received amphotericin B in 5% glucose infusion for 6 h and group B for 2 h. Amphotericin B was given at an initial dosage of 0.05 mg/kg body weight, increasing to 1 mg/kg body weight on day 5, and this dose was given daily until patients had received a total of 20 mg/kg body weight. Toxicities like rise in serum creatinine, fall in serum potassium levels and diminution in appetite were no different in the two groups. There was no difference in clinical, parasitological and ultimate cures between the two groups. Man-hours spent in treating patients in the 2-h infusion group were one-third of the time with the 6-h infusion group. It was concluded that 2-h infusion of amphotericin B caused fewer infusion-related toxicities (shivering and fever) and saved time spent in treating patients; hence it should be recommended for use.

Comparison of three treatment regimens with liposomal amphotericin B (AmBisome) for visceral leishmaniasis in India: a randomized dose-finding study.

The efficacy and safety of 3 regimens of liposomal amphotericin B (AmBisome) in the treatment of Indian visceral leishmaniasis were compared in a prospective open randomized trial. Thirty parasitologically confirmed patients were randomly divided into 3 equal treatment groups; group 1 received AmBisome 2mg/kg on days 1, 2, 3, 4, 5, 6, and 10 (total dose 14 mg/kg); group 2 received AmBisome 2 mg/kg on days 1, 2, 3, 4, and 10 (total dose 10 mg/kg); group 3 received the same dosage on 1, 5 and 10 (total dose 6 mg/kg). Clinical cure resulted in all patients by day 24. Haemoglobin, white blood cell count, body weight and serum albumin level improved on day 24 and became normal by day 180. No patient relapsed within 12 months of follow-up. Side effects were minimal. One patient in group 2 died after 2 months from an unrelated disease. Liposomal amphotericin B is a promising new drug which is highly efficacious in the treatment of Indian kala-azar and produces minimal toxicity.

A prospective randomized, comparative, open-label trial of the safety and efficacy of paromomycin (aminosidine) plus sodium stibogluconate versus sodium stibogluconate alone for the treatment of visceral leishmaniasis.

Response to treatment with organic pentavalent antimonials, the standard first-line treatment for visceral leishmaniasis (VL), has been decreasing since their introduction into India. Combining sodium stibogluconate (SB) with paromomycin (PM) may be an efficient alternative to single-agent therapy. This trial was designed to assess the safety and efficacy of PM 12 or 18 mg/kg daily plus SB 20 mg/kg daily for 21 days compared to SB alone for 30 days. One hundred and fifty patients were randomly assigned in 1996 to 1 of the 3 treatments and followed-up for 180 days. At the end of treatment, 49 of 52 patients receiving PM12 + SB, 46 of 48 receiving PM18 + SB, and 27 of 49 patients receiving SB alone, were cured. During follow-up there was 1 relapse in each of the treatment groups, giving final cure rates of 48 of 52 (92.3%) for PM12 + SB, 45 of 48 (93.8%) for PM18 + SB, and 26 of 49 (53.1%) for SB. PM plus SB for 21 days at either 12 or 18 mg/kg daily was significantly more effective than SB alone for 30 days (chi 2 P < 0.001). One patient (SB alone) had experienced a serious adverse event: cardiotoxicity at day 8 (myocarditis and ECG changes) which caused withdrawal from the study. Only 19 of 100 patients enrolled in the PM treatment arms had a complete audiogram series conducted thus making it difficult to assess oto-toxicity. PM 12 or 18 mg/kg daily plus a standard dose of SB for 21 days was statistically more effective than SB in producing a final cure for patients with VL in Bihar, India.

Broadband dielectric spectroscopy study of molecular dynamics in the glass-forming liquid crystal isopentylcyanobiphenyl dispersed with aerosils.

The glass-forming liquid crystal isopentylcyanobiphenyl (CB15) filled with different concentrations of hydrophilic and hydrophobic aerosils has been investigated by broadband dielectric spectroscopy in the frequency range from 10(-2) Hz to 10(7) Hz over a temperature range of 173 K-300 K. CB15 that consists of chiral molecules has a monotropic system of phases nematic (N*) and smectic-A upon supercooling and forms a glass further on. In the isotropic phase a single Davidson-Cole process exists in the substance, which is due to the rotation of the molecules around their short axes. In the supercooled N* phase a Cole-Cole process that is an order of magnitude faster is additionally present and is due to the rotation in a cone around the local director. The relaxation times of the process due to rotation around short axes obey the empirical Vogel-Fulcher-Tamman behavior typical for glass-forming systems. Filling of the liquid crystal (LC) with different concentrations of hydrophilic aerosils leads to the emergence of a slow relaxation process that grows with the increasing concentration of the aerosils. The aerosil particles, which form a three-dimensional network dividing the LC phase into domains, have little effect on the relaxation times of the bulk processes. As a consequence the glass transition temperature is merely affected. On the other hand, in LCs dispersed with hydrophobic aerosils the slow process is quite weak. The slow process is attributed to the relaxation of the molecules that are homeotropically attached at the surfaces of the aerosil particles. The LC-aerosil surface interaction leads to a considerable slowing down of the molecular rotation around their short axis. The process has an Arrhenius-like temperature dependence of the relaxation times with an activated type of dynamics, which can be explained by considering a nonincreasing rearranging region of cooperativity in surface layers.