Performance analysis of hybrid deep learning framework using a vision transformer and convolutional neural network for handwritten digit recognition

Digitization created a demand for highly efficient handwritten document recognition systems. A handwritten document consists of digits, text, symbols, diagrams, etc. Digits are an essential element of handwritten documents. Accurate recognition of handwritten digits is vital for effective communication and data analysis. Various researchers have attempted to address this issue with modern convolutional neural network (CNN) techniques. Even after training, CNN filter weights remain unchanged despite the high identification accuracy. As a result, the process cannot flexibly adapt to input changes. Hence computer vision researchers have recently become interested in Vision Transformers (ViTs) and Multilayer Perceptrons (MLPs). The shortcomings of CNNs gave rise to a hybrid model revolution that combines the best elements of the two fields. This paper analyzes how the hybrid convolutional ViT model affects the ability to recognize handwritten digits. Also, the real-time data contains noise, distortions, and varying writing styles. Hence, cleaned and uncleaned handwritten digit images are used for evaluation in this paper. The accuracy of the proposed method is compared with the state-of-the-art techniques, and the result shows that the proposed model achieves the highest recognition accuracy. Also, the probable solutions for recognizing other aspects of handwritten documents are discussed in this paper.•Analyzed the effect of convolutional vision transformer on cleaned and real-time handwritten digit images.•The model's performance improved with the implication of cross-validation and hyper-parameter tuning.•The results show that the proposed model is robust, feasible, and effective on cleaned and uncleaned handwritten digits.

Enhancement of handwritten text recognition using AI-based hybrid approach

Handwritten text recognition (HTR) within computer vision and image processing stands as a prominent and challenging research domain, holding significant implications for diverse applications. Among these, it finds usefulness in reading bank checks, prescriptions, and deciphering characters on various forms. Optical character recognition (OCR) technology, specifically tailored for handwritten documents, plays a pivotal role in translating characters from a range of file formats, encompassing both word and image documents. Challenges in HTR encompass intricate layout designs, varied handwriting styles, limited datasets, and less accuracy achieved. Recent advancements in Deep Learning and Machine Learning algorithms, coupled with the vast repositories of unprocessed data, have propelled researchers to achieve remarkable progress in HTR. This paper aims to address the challenges in handwritten text recognition by proposing a hybrid approach. The primary objective is to enhance the accuracy of recognizing handwritten text from images. Through the integration of Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) with a Connectionist Temporal Classification (CTC) decoder, the results indicate substantial improvement. The proposed hybrid model achieved an impressive 98.50% and 98.80% accuracy on the IAM and RIMES datasets, respectively. This underscores the potential and efficacy of the consecutive use of these advanced neural network architectures in enhancing handwritten text recognition accuracy. •The proposed method introduces a hybrid approach for handwritten text recognition, employing CNN and BiLSTM with CTC decoder.•Results showcase a remarkable accuracy improvement of 98.50% and 98.80% on IAM and RIMES datasets, emphasizing the potential of this model for enhanced accuracy in recognizing handwritten text from images.

Anomaly-informed remaining useful life estimation (AIRULE) of bearing machinery using deep learning framework

A rolling bearing is a crucial element within rotating machinery, and its smooth operation profoundly influences the overall well-being of the equipment. Consequently, analyzing its operational condition is crucial to prevent production losses or, in extreme cases, potential fatalities due to catastrophic failures. Accurate estimates of the Remaining Useful Life (RUL) of rolling bearings ensure manufacturing safety while also leading to cost savings.•This paper proposes an intelligent deep learning-based framework for remaining useful life estimation of bearings on the basis of informed detection of anomalies.•The paper demonstrates the setup of an experimental bearing test rig and the collection of bearing condition monitoring data such as vibration data.•Advanced hybrid models of Encoder-Decoder LSTM demonstrate high forecasting accuracy in RUL estimation.

Temporal and institutional trends in robotic surgery

Robotic surgery has become increasingly prevalent in general surgery practice. While previous studies have shown the safety and efficacy of robotic assistance in laparoscopic general surgery procedures, few studies have evaluated the temporal and regional trends in implementation. In our retrospective population-based study, we aim to evaluate the national trends in robotic surgery. National Inpatient Database (NIS 2009-2014) was used to identify adults who underwent robotic assisted surgery (ICD 9 codes 17.41 to 17.49). Robotic procedures related to seven abdominal organ systems were compared against the trends of Urology, Gynaecologic, and Orthopedic robotic procedures. Discharge weights were applied to calculate National temporal trends separated by hospital size, teaching status and US geographic region. 894,163 patients received a robotic assisted procedure between 2009 and 2014 with 64% increase in utilization. The largest percent change was witnessed in biliary robotic procedures with 2984% change in utilization, followed by hernia (1376%). Lowest percent change was witnessed in esophageal procedures with 114% increase. Medium sized hospitals had the largest change in robotic utilization (41%), with large institutions seeing 18% decrease. Gastric procedures were the most common robotic procedure performed at small institutions (7917 total cases; 316%). Large institutions saw an overall decrease in gastric (- 47%), esophageal (- 17%), small and large intestinal (-16%), and hepatic (- 7%) robotic procedures. Rural non-teaching hospitals saw the largest increase in robotic surgery (274%). Urban non-teaching hospitals saw a decrease of 29%. While urban teaching institutions saw a 20% and 6% increase in gynecological and urological procedures, an overall decrease was seen in esophageal (- 10%), gastric (- 12%), intestinal (- 11%), hepatic (- 17%), biliary (- 10%), pancreatic (- 11%) and hernia procedures (- 14%). Biliary procedures saw the largest increase in rural institutions (740 cases; 392%), followed by hernia (144% increase). South region of the nation had the largest increase in robotic procedures (23%). No change was seen in the use of robotic surgery in the northeast region with the midwest and west seeing an overall decrease (- 4% and - 22%, respectively). Our study highlights the increase in use of robotics for both general and specialty surgery, with an increase in utilization over time. Increased incidence of robotic surgery in smaller, rural institutions with overall decrease in larger, urban teaching hospitals suggests increasing comfort in robotic surgery in the community setting. Further studies are necessary to evaluate the factors associated with increased utilization in smaller institutions.

Knockdown of Tlr3 in dorsal striatum reduces ethanol consumption and acute functional tolerance in male mice

Systemic activation of toll-like receptor 3 (TLR3) signaling using poly(I:C), a TLR3 agonist, drives ethanol consumption in several rodent models, while global knockout of Tlr3 reduces drinking in C57BL/6J male mice. To determine if brain TLR3 pathways are involved in drinking behavior, we used CRISPR/Cas9 genome editing to generate a Tlr3 floxed (Tlr3F/F) mouse line. After sequence confirmation and functional validation of Tlr3 brain transcripts, we injected Tlr3F/F male mice with an adeno-associated virus expressing Cre recombinase (AAV5-CMV-Cre-GFP) to knockdown Tlr3 in the medial prefrontal cortex, nucleus accumbens, or dorsal striatum (DS). Only Tlr3 knockdown in the DS decreased two-bottle choice, every-other-day (2BC-EOD) ethanol consumption. DS-specific deletion of Tlr3 also increased intoxication and prevented acute functional tolerance to ethanol. In contrast, poly(I:C)-induced activation of TLR3 signaling decreased intoxication in male C57BL/6J mice, consistent with its ability to increase 2BC-EOD ethanol consumption in these mice. We also found that TLR3 was highly colocalized with DS neurons. AAV5-Cre transfection occurred predominantly in neurons, but there was minimal transfection in astrocytes and microglia. Collectively, our previous and current studies show that activating or inhibiting TLR3 signaling produces opposite effects on acute responses to ethanol and on ethanol consumption. While previous studies, however, used global knockout or systemic TLR3 activation (which alter peripheral and brain innate immune responses), the current results provide new evidence that brain TLR3 signaling regulates ethanol drinking. We propose that activation of TLR3 signaling in DS neurons increases ethanol consumption and that a striatal TLR3 pathway is a potential target to reduce excessive drinking.

Clinical effects of Lactobacillus reuteri probiotic in chronic periodontitis - a systematic review

OBJECTIVE

This systematic review examines the effectiveness of Lactobacillus reuteri as an adjunct to scaling and root planing in the treatment of chronic periodontitis.

MATERIALS AND METHODS

Scopus, PubMed, and Web of Science databases were searched according to specific inclusion and exclusion criteria in October 2022. Randomized control trials that evaluated the effects of Lactobacillus reuteri in patients with periodontitis were included. The primary outcome was pocket depth and clinical attachment levels, while the secondary outcome considered was bleeding on probing, microbial levels, and gingival index score. Study quality was assessed based on the Cochrane Handbook for Systematic Reviews of Interventions and the ROB2 tool.

RESULTS

A total of eleven studies that examined 369 subjects were included in the review. Adults in the age group of 18-70 years of age suffering from chronic periodontitis were evaluated. Eight out of the eleven studies reported statistically significant improvement in the intergroup pocket depths, whereas seven studies showed a statistically significant reduction in the clinical attachment levels in the probiotic group. Three studies showed no significant improvement in the pocket depth levels in the probiotic group as compared to the controls. Four studies showed no significant reduction in clinical attachment levels between the two groups. The overall risk of bias was high in four studies, while seven studies reported some concerns about the risk of bias.

CONCLUSIONS

Based on the limited evidence available, the adjunctive use of Lactobacillus reuteri to scaling and root planing may provide some additional benefit in improving periodontal parameters.

Multimodal Abstractive Summarization using bidirectional encoder representations from transformers with attention mechanism

In recent decades, abstractive text summarization using multimodal input has attracted many researchers due to the capability of gathering information from various sources to create a concise summary. However, the existing methodologies based on multimodal summarization provide only a summary for the short videos and poor results for the lengthy videos. To address the aforementioned issues, this research presented the Multimodal Abstractive Summarization using Bidirectional Encoder Representations from Transformers (MAS-BERT) with an attention mechanism. The purpose of the video summarization is to increase the speed of searching for a large collection of videos so that the users can quickly decide whether the video is relevant or not by reading the summary. Initially, the data is obtained from the publicly available How2 dataset and is encoded using the Bidirectional Gated Recurrent Unit (Bi-GRU) encoder and the Long Short Term Memory (LSTM) encoder. The textual data which is embedded in the embedding layer is encoded using a bidirectional GRU encoder and the features with audio and video data are encoded with LSTM encoder. After this, BERT based attention mechanism is used to combine the modalities and finally, the BI-GRU based decoder is used for summarizing the multimodalities. The results obtained through the experiments that show the proposed MAS-BERT has achieved a better result of 60.2 for Rouge-1 whereas, the existing Decoder-only Multimodal Transformer (D-MmT) and the Factorized Multimodal Transformer based Decoder Only Language model (FLORAL) has achieved 49.58 and 56.89 respectively. Our work facilitates users by providing better contextual information and user experience and would help video-sharing platforms for customer retention by allowing users to search for relevant videos by looking at its summary.

A comparison of deep transfer learning backbone architecture techniques for printed text detection of different font styles from unstructured documents

Object detection methods based on deep learning have been used in a variety of sectors including banking, healthcare, e-governance, and academia. In recent years, there has been a lot of attention paid to research endeavors made towards text detection and recognition from different scenesor images of unstructured document processing. The article's novelty lies in the detailed discussion and implementation of the various transfer learning-based different backbone architectures for printed text recognition. In this research article, the authors compared the ResNet50, ResNet50V2, ResNet152V2, Inception, Xception, and VGG19 backbone architectures with preprocessing techniques as data resizing, normalization, and noise removal on a standard OCR Kaggle dataset. Further, the top three backbone architectures selected based on the accuracy achieved and then hyper parameter tunning has been performed to achieve more accurate results. Xception performed well compared with the ResNet, Inception, VGG19, MobileNet architectures by achieving high evaluation scores with accuracy (98.90%) and min loss (0.19). As per existing research in this domain, until now, transfer learning-based backbone architectures that have been used on printed or handwritten data recognition are not well represented in literature. We split the total dataset into 80 percent for training and 20 percent for testing purpose and then into different backbone architecture models with the same number of epochs, and found that the Xception architecture achieved higher accuracy than the others. In addition, the ResNet50V2 model gave us higher accuracy (96.92%) than the ResNet152V2 model (96.34%).

Advances in the co-production of biosurfactant and other biomolecules: statistical approaches for process optimization

An efficient microbial conversion for simultaneous synthesis of multiple high-value compounds, such as biosurfactants and enzymes, is one of the most promising aspects for an economical bioprocess leading to a marked reduction in production cost. Although biosurfactant and enzyme production separately have been much explored, there are limited reports on the predictions and optimization studies on simultaneous production of biosurfactants and other industrially important enzymes, including lipase, protease, and amylase. Enzymes are suited for an integrated production process with biosurfactants as multiple common industrial processes and applications are catalysed by these molecules. However, the complexity in microbial metabolism complicates the production process. This study details the work done on biosurfactant and enzyme co-production and explores the application and scope of various statistical tools and methodologies in this area of research. The use of advanced computational tools is yet to be explored for the optimization of downstream strategies in the co-production process. Given the complexity of the co-production process and with various new methodologies based on artificial intelligence (AI) being invented, the scope of AI in shaping the biosurfactant-enzyme co-production process is immense and would lead to not only efficient and rapid optimization, but economical extraction of multiple biomolecules as well.

Comparative Evaluation of Relapse After LeFort 1 Osteotomy in Cleft Palate and Non-cleft Individuals: A Longitudinal Study

PURPOSE

The aim of this study was to assess and compare the likelihood of relapse one year after LeFort I advancement surgery in patients with and without cleft lip and palate.

METHODS

A retrospective observational study which included two groups of participants who underwent LeFort I maxillary advancement was performed. Group 1 included 10 non-cleft subjects and Group 2 included 21 subjects with cleft palate. These maxillary deficient patients were chosen and operated using a technique where only a sagittal displacement was intended. Patients who underwent additional mandibular surgery, significant vertical or transverse alterations, or both were excluded. Pre-operative (T1), immediately post-operative (T2), and minimum one-year follow-up (T3) lateral cephalograms were studied for each group. Skeletal stability and dental stability after LeFort I surgery at a minimum of one-year follow-up in cleft palate and non-cleft patients were evaluated.

RESULTS

For the given sample size, relapse tendencies showed statistically significant differences between cleft palate patients and non-cleft palate patients after maxillary advancement. The sella nasion angle and horizontal overlap of the maxillary and mandibular incisors (overjet) decreased by 2 degrees and 0.9 mm respectively in the cleft palate group while decreasing by 1.10 degrees and 0.40 mm in the non-cleft group.

CONCLUSIONS

After maxillary advancement with LeFort I osteotomy and miniplate fixation in patients with cleft palate and non-cleft patients, some degree of relapse was detected in both groups for the given sample size after one year post-operatively. The cleft palate group displayed additional relapse tendencies when compared to the non-cleft group.

A Comparative Evaluation of Three Methods of Disinfection of Gypsum Casts and the Changes in Surface Roughness and Dimensional Accuracy after Disinfection-An Ex Vivo Study

BACKGROUND

One of the major problems of everyday dental practice is cross-contamination. It can place office personnel, dentists, and patients at risk of acquiring serious illness. Disinfection helps in controlling this cross-contamination to an extent. The evaluation was done to find the efficient disinfection method on gypsum casts.

AIMS

The aim of this study is to evaluate and compare the efficacy of three methods of disinfection of gypsum casts, namely, chemical disinfection by immersion, spray method, and microwave method, and also to evaluate and compare changes in surface roughness and dimensional accuracy between the three methods after disinfection.

MATERIALS AND METHODS

Ex vivo and experimental study. Thirty participants were selected, and impressions of the maxillary arch were made using polyvinyl siloxane impression material. Ninety type IV die stone gypsum casts were poured. It was divided into three groups and was subjected to chemical disinfection by immersion and spray methods, and microwave method. The disinfected casts were evaluated for microbial growth, surface roughness, and dimensional accuracy. It was performed by using the one-way analysis of variance test and paired t-test followed by the Kruskal - Wallis test and Wilcoxon signed rank test (α = 0.05).

RESULTS

Microwave disinfection was more effective than both immersion and spray chemical disinfection methods (P < 0.010 and <0.001). The surface roughness of the microwave-irradiated casts had significantly increased after disinfection. However, there were no significant dimensional changes by any of the methods of disinfection.

CONCLUSION

Within the limitations of the study, the microwave method of disinfection is more effective in eradicating microorganisms when compared to chemical methods of disinfection by immersion and spray methods.

Validation of Diagnostic Utility of Salivary Urea in Patients with Chronic Kidney Disease in Chennai: A Cross-Sectional Study

BACKGROUND

Serum urea levels are essential for the diagnosis of chronic kidney disease (CKD), as they are a measure of renal function. Salivary urea has been used as an alternative to serum urea in patients with CKD.

AIM

To estimate serum urea levels and their correlation with salivary urea levels in order to validate the use of saliva as an alternative diagnostic tool for renal dysfunction.

PATIENTS AND METHODS

Thirty healthy individuals and 150 patients with CKD (30 from each CKD stage) were recruited for the study. Serum and salivary samples were collected to estimate the urea levels, and statistical analyses were performed.

RESULTS

Statistically significant (P < 0.05) differences in the mean serum and salivary urea levels were observed between the healthy and each stage of the CKD group; significance at P < 0.05 was observed within the stages of CKD as well. Correlation analysis also showed significant differences between the serum and salivary urea levels at each stage of the disease. Receiver operating characteristic analysis revealed a higher area under the curve of 0.917 at a 95% confidence interval with a salivary urea cut-off of 28.25 mg/dl (sensitivity and specificity: 88% and 84%, respectively).

CONCLUSION

The variability in salivary urea levels in the early and late stages suggests the use of salivary urea in the late stages of CKD. It is suggested that salivary urea levels may be used as an alternative to serum to assess and monitor the progression of renal impairment along with other standard renal function markers.

Comparative Evaluation of Various Lens and Ring Flash Combination for Intraoral Photography

BACKGROUND

Extraoral and intraoral dental photographs serve as preoperative records and document the entire treatment. Correctly composed orthodontic photographs are crucial for remote diagnosis and may serve as a bulwark against medicolegal challenges.

MATERIALS AND METHODS

In this prospective study, intraoral frontal photographs of patients with ideal occlusion were taken using two types of lenses (EF-S 18-55 mm f/3.5-5.6 IS STM lens (Canon, Tokyo, JP), SP 90 mm F/2.8 MACRO VC lens (Model F017 Tamron, NY, USA)) and two different ring flash systems (Meike FC-100 Macro Ring LED Light (Meike, China), Macro Ring flash Lite YN-14EX (Yongnuo digital, China)). The combination of lens and flash used was grouped into four groups. Twenty-eight intraoral photographs of patients were taken. An image quality assessment survey was distributed among two groups - 50 orthodontists and 50 other dental specialists.

RESULTS

The participants were asked to assess all the intraoral images and subjectively score them on a scale of one to ten, with one being very poor and ten being excellent, considering the sharpness, color, brightness, contrast, and overall quality of the image. The general dentists rated the images taken with a 90-mm macro lens and ring flash as the best quality photographs. Images obtained using an 18-55 mm lens and ring LED received significantly lesser scores and were graded good by dentists.

CONCLUSION

This combination of lens and flash may prove a valuable investment in the long-term aiding in excellent dental images for diagnosis and treatment monitoring.

Artificial intelligence: Machine learning approach for screening large database and drug discovery

Recent research in drug discovery dealing with many faces difficulties, including development of new drugs during disease outbreak and drug resistance due to rapidly accumulating mutations. Virtual screening is the most widely used method in computer aided drug discovery. It has a prominent ability in screening drug targets from large molecular databases. Recently, a number of web servers have developed for quickly screening publicly accessible chemical databases. In a nutshell, deep learning algorithms and artificial neural networks have modernised the field. Several drug discovery processes have used machine learning and deep learning algorithms, including peptide synthesis, structure-based virtual screening, ligand-based virtual screening, toxicity prediction, drug monitoring and release, pharmacophore modelling, quantitative structure-activity relationship, drug repositioning, polypharmacology, and physiochemical activity. Although there are presently a wide variety of data-driven AI/ML tools available, the majority of these tools have, up to this point, been developed in the context of non-communicable diseases like cancer, and a number of obstacles have prevented the translation of these tools to the discovery of treatments against infectious diseases. In this review various aspects of AI and ML in virtual screening of large databases were discussed. Here, with an emphasis on antivirals as well as other disease, offers a perspective on the advantages, drawbacks, and hazards of AI/ML techniques in the search for innovative treatments.

Diode laser targeting red-complex bacteria in periodontitis: a systematic review

OBJECTIVE

This systematic review examines the effectiveness of diode laser irradiation in reducing the levels of red complex bacteria as well as periodontal parameters of pocket depth and clinical attachment level.

MATERIALS AND METHODS

We conducted electronic searches across databases such as Scopus, Embase, Medline, and Web of Science databases in July 2022. Randomized controlled trials that evaluated the reduction of red-complex bacteria in patients with periodontitis using diode lasers were included. The primary focus was the reduction in the microbial count of red complex bacteria, whereas probing depth and attachment level were considered secondary outcomes. Articles in languages other than English were excluded. Study quality was assessed based on the Cochrane Handbook for Systematic Reviews of Interventions and the ROB2 tool.

RESULTS

After searching the databases, eight independent studies were included, with a sample size of 210 subjects. The average age group of the study population was 30-60 years, and there was a lack of consensus on the antimicrobial effect of diode lasers. Out of the eight studies, four studies reported no significant difference in the levels of red complex bacteria before and after laser application. Three studies reported significantly lower levels of red complex bacteria in the intergroup comparison. One study reported that laser had no significant effect on intergroup bacterial levels. The combination of diode laser irradiation with scaling reduced the count of red complex bacteria and improved the clinical parameters, although not significantly.

CONCLUSIONS

Based on the limited evidence available, the adjunctive use of diode laser for scaling and root planning may provide some additional benefit in terms of reduction of red complex bacterial count and clinical parameters. Further well-designed trials and the use of objective measures are necessary before outlining universal guidelines for best practice. The adjunctive use of diode laser in non-surgical periodontal therapy may provide a reduction in the red complex microbial count and improvement in clinical parameters, decreasing the need for periodontal surgery.

Public Awareness on Perio-systemic Interplay: A Cross-Sectional Survey in South India

BACKGROUND

The oral cavity has been referred to as "the gateway to overall health." It is also said to be the meeting point of medicine and dentistry.

AIMS

Our study sought to determine the extent to which the public was aware of the connection between oral/periodontal conditions and general health.

SETTINGS AND DESIGN

The observational cross-sectional study's questionnaire was sectioned into oral health awareness, systemic influence on oral health, and personal oral health assessment.

MATERIALS AND METHODS

A total of 994 responses were recorded and a Chi-square test was performed to uncover the relationships using SPSS version 22.0. According to responses, 70% of the population on average comprehended the responses to the majority of the oral health awareness-related questions.

RESULTS

It has been noticed that only 30% of the general public was aware of the prevalent health issues like diabetes, hypertension, and malnutrition's impact on dental health. However, more than 60% had confidence in their oral health and gave a rating of at least 5.

CONCLUSION

The study indicates that a good number of the population was prioritizing their oral health. However, there exists a definitive need to improve oral health awareness thereby ameliorating the overall health of an individual.

Correlation of clinical features with neuroimaging in children with mild head injury

Objectives

Mild head injury is defined as a pediatric Glasgow Coma Scale (GCS) score of 13-15 at admission following head trauma. There are no clear indications for neuroimaging in such children. The aim of our study was to analyze the correlation of symptoms commonly encountered following mild head injury with any abnormality on the computed tomography (CT) scan.

Materials and Methods

This is a retrospective and observational study done in a tertiary care hospital. Records of all the children fulfilling the inclusion criteria were retrieved. Demographic details such as age, gender, and type of injury (hit by a blunt object, fall from height, and road traffic accident), symptoms such as presence and number of episodes of vomiting, presence and duration of loss of consciousness (LOC), presence of drowsiness, headache, giddiness, seizures, focal deficits and the GCS score, and CT scan findings were noted in a predesigned proforma. A CT scan with any of the following abnormalities such as skull fracture, cerebral edema, intracranial bleed (epidural hematoma, subdural hematoma, intracerebral hematoma, and subarachnoid hemorrhage), and cerebral contusion was considered as abnormal. Any neurosurgical intervention done was also noted.

Results

A total of 134 children in the age group of 1 month to 18 years with mild head injury were included in this study. The median (interquartile range) age of the children was 5 (2, 8) years, with majority being males. Road traffic accidents (34.3%) were the most common cause of injury, followed by fall from lesser than 3 feet height (31.3%). The most common symptoms were vomiting (43.6%), scalp/facial abrasions (37.2%), and LOC (31.9%). CT scan was abnormal in 53.7% of the cases, with skull fracture (35.1%) being the most common finding, followed by cerebral edema (13.4%). Among all the symptoms, ear/nosebleed or cerebrospinal fluid (CSF) otorrhea/rhinorrhea had a statistically significant association with a positive CT scan with P = 0.05 and an odds ratio of 1.4 (95% confidence interval, 1-1.9).

Conclusion

Children with mild head injury presenting with clinical features such as bleeding from the ear or nose and CSF otorrhea/rhinorrhea are more likely to have an abnormal CT scan. Hence, such children require close neurological observation and prompt intervention if necessary. However, abnormality on CT scan cannot be reliably ruled out based on the symptoms alone.

Evaluation of scanning accuracy for two commercially available intraoral scanners in reproducing orthodontic bracket dimensions

OBJECTIVE

The aim of this study was to compare the scanning accuracy of two different intra-oral scanners- MEDIT i500 and TRIOS 3 shape in reproducing orthodontic bracket dimensions.

MATERIALS AND METHODS

This in vivo cross-sectional study comprised seven subjects with a full complement of permanent dentition without third molars. Complete arch scanning was carried out with two intraoral scanners, such as MEDIT i500, TRIOS 3 shape, after bonding with brackets. The control group consisted of bracket dimensions measured directly by using Vernier calipers before bonding. Bracket dimensions of three-dimensional (3D) images were measured by using OrthoAnalyzer software. The accuracy of intraoral scanning was investigated by comparing bracket dimensions among the three groups using One-way ANOVA and Post-Hoc Tukey HSD test, and by evaluating outcomes for each quadrant and an individual tooth in complete-arch scans.

RESULTS

When comparing bracket dimensions of 3D images with manual measurements using a traditional Vernier caliper, MEDIT i500 showed no significant difference when compared to the control group (p>0.05) in full arch scanning as well as the quadrant and single tooth scans in complete arch scanning. TRIOS 3 shape showed a significant difference when compared to the control and MEDIT i500 group (p<0.05) for all parameters.

CONCLUSIONS

MEDIT i500 showed higher accuracy and better reproduction of orthodontic bracket dimensions than TRIOS 3 shape.

Comparison of bacterial colonization on absorbable non-coated suture with Triclosan- or Chlorhexidine-coated sutures: a randomized controlled study

OBJECTIVE

The oral cavity is a colossal reservoir for the bacteria. The healing of tissues is compromised after flap surgery, particularly in the presence of sutures, as they can act as repositories for bacteria, ultimately leading to surgical site infections. Hence, antibacterial-coated sutures have been considered as an alternative to reduce the risk of these infections and further improve the wound healing of the tissues after flap surgery. Since minimal information is available on the effect of antibacterial-coated sutures on periodontal tissues, this study aims to clinically and microbiologically assess the antibacterial efficacy of Triclosan (TCS) and Chlorhexidine-coated sutures (CCS) on periodontal tissues compared to non-coated sutures (NCS).

PATIENTS AND METHODS

A total of 75 subjects with moderate to severe periodontitis were included in the study and randomly allocated to one of the three groups, (TCS, CCS, and NCS groups) equally. Suture removal was performed on postoperative day 8, and parameters such as wound healing and post-operative pain were evaluated. The retrieved suture samples were subjected to microbiological analysis and the bacteria were identified quantitatively and qualitatively.

RESULTS

Intragroup analysis of the wound healing index and post-operative pain for all the groups showed a significant improvement (p<0.01), from day 8 to day 30. Intergroup analysis of the wound healing index revealed significant wound healing (p<0.05) on day 15 and day 30. For post-operative pain, intergroup analyses showed significantly low pain scores (p<0.01) for the TCS group. Microbiologic analysis of aerobic colony counts in both anterior and posterior regions revealed significantly (p<0.01) least colony counts in TCS and highest colony counts in NCS groups, respectively. Although anaerobic colony counts were not statistically significant, relatively fewer colony counts were identified in the TCS group. Whereas, relatively higher anaerobic colony counts were seen in the CCS group in the anterior region and in the NCS group in the posterior region. Qualitative assessment revealed higher amounts of Streptococcus and Staphylococcus species in all the three groups (TCS, CCS, and NCS groups).

CONCLUSIONS

Antibacterial-coated sutures, particularly Triclosan-coated sutures, are effective in reducing bacterial accumulation compared to non-coated sutures. Therefore, these sutures can be effectively utilized in periodontal flap surgery.

Success rate of immediately loaded implants in the posterior zone

Dental implants are considered an ideal treatment for a missing single tooth. Immediate loading of implants can hasten the procedure, providing comfort to the patients. Recently, immediate loading of implants has gained much importance as it helps hasten the procedure and provides more comfort to patients. A previous systematic review published 5 years ago compared the success rates between immediate and conventional loading. There are several factors that influence the success rate of implants that were not discussed in detail in the previous review. Hence, the present systematic review is done to report differences in the outcomes from single implant restorations of missing teeth in the posterior region in patients who were subjected to immediate loading and conventional loading. A follow up for 1 year was done. Electronic databases of Medline, Scopus, and Web of Science were searched for publications in the English Language during May 2021. The search results yielded 306 articles, out of which 225 were excluded based on title and abstract screening. Screening of the remaining 81 full text articles yielded 14 original research articles that satisfied the predefined inclusion criteria. Meta analysis was not possible due to the heterogeneity of the data. The overall success rate of the immediate loading of a single implant is 94.31%. Implants in the maxillary region had a higher survival rate than those in the mandibular region. The age range between 18 and 80 years showed good prognosis and outcomes in older individuals. Good oral hygiene was emphasized for all patients to prevent any secondary conditions or delays in healing.

Development of a three-dimensional organoid model to explore early retinal phenotypes associated with Alzheimer's disease

Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by the accumulation of Aβ plaques and neurofibrillary tangles, resulting in synaptic loss and neurodegeneration. The retina is an extension of the central nervous system within the eye, sharing many structural similarities with the brain, and previous studies have observed AD-related phenotypes within the retina. Three-dimensional retinal organoids differentiated from human pluripotent stem cells (hPSCs) can effectively model some of the earliest manifestations of disease states, yet early AD-associated phenotypes have not yet been examined. Thus, the current study focused upon the differentiation of hPSCs into retinal organoids for the analysis of early AD-associated alterations. Results demonstrated the robust differentiation of retinal organoids from both familial AD and unaffected control cell lines, with familial AD retinal organoids exhibiting a significant increase in the Aβ42:Aβ40 ratio as well as phosphorylated Tau protein, characteristic of AD pathology. Further, transcriptional analyses demonstrated the differential expression of many genes and cellular pathways, including those associated with synaptic dysfunction. Taken together, the current study demonstrates the ability of retinal organoids to serve as a powerful model for the identification of some of the earliest retinal alterations associated with AD.

In Vitro Evaluation of marginal adaptation of polyether ether ketone and zirconia copings

Background

Polyether ether ketone (PEEK) has emerged as a new thermoplastic material with potential applications as a restorative material. Aim: This study aimed to evaluate the marginal adaptation of PEEK copings compared to zirconia copings using field emission scanning electron microscopy.

Materials and Methods

A freshly extracted maxillary central incisor was prepared for a full-coverage restoration following standard principles of tooth preparation. The tooth was sent to a laboratory for fabrication of samples using computer-aided design and manufacturing (CAD/CAM). Twenty samples of polyether ether ketone (PEEK) copings (group A) and 20 of zirconia copings were fabricated (group B). The copings were scanned under a field emission scanning electron microscope and measurements were taken at four distinct points. The marginal adaptation over the buccal, lingual, mesial, and distal margins for both groups was evaluated. One-way analysis of variance (ANOVA) and independent t test were applied.

Results

Our findings indicate that PEEK showed better marginal adaptation than zirconia at all measurement points. The mean marginal gap value of the PEEK group was 33.99 ± 8.81 μm and of the zirconia group was 56.21 ± 15.07 μm. On comparing marginal adaptation among the mesial, distal, buccal, and lingual aspects, PEEK showed better adaptation on all four margins, with the best adaptation on the buccal margin that had the lowest mean gap value of 29.27 ± 6.07 μm. The zirconia group adapted best at the distal margin, with a lowest mean gap value of 53.58 ± 15.25 μm (P ≤ 0.05).

Conclusion

PEEK copings had better marginal adaptation and fit compared to zirconia copings. It may have applications as a restorative material in fixed prostheses.

Root surface changes following manual and ultrasonic instrumentation - a scanning electron microscopic study

OBJECTIVE

The aim of this study is to assess the efficacy of hand and ultrasonic scaling and to evaluate the surface roughness on the root surface of periodontally involved teeth using a scanning electron microscope.

PATIENTS AND METHODS

A sample of 90 single-rooted teeth with a hopeless prognosis was selected for the study and divided into three separate groups. Group I consist of no treatment. In Group II, hand scaling was done using Gracey curettes, and in Group III, ultrasonic scaling was done. The teeth were then extracted and fixed in 10% formaldehyde solution for 24-48 hours and subjected to scanning electron microscopic (SEM) evaluation.

RESULTS

The SEM analysis revealed that the remaining calculus index was found to be similar in the ultrasonic group and the hand scaling group, whereas the surface roughness was found to be the least in the ultrasonic group.

CONCLUSIONS

Hand instrumentation has resulted in more surface roughness as compared to ultrasonic instruments.

A split-mouth randomized controlled trial to compare the rate of canine retraction after a soft tissue procedure compared against a corticotomy procedure for accelerated tooth movement

Background and Aim

Various methods to accelerate the orthodontic tooth movement have been used, among which corticotomy is considered to be the most common one. The suggested reasoning for such acceleration was the regional acceleratory phenomenon (RAP). Since the RAP is a property of both the hard and soft tissues, we designed a soft tissue flap procedure to compare the effects with the conventional corticotomy procedure. A split-mouth study was conducted where the two procedures were assessed in a single participant.

Patients and Methods

The total sample size was calculated to be 40 with 20 participants in each group. The rate of tooth movement was the primary outcome measure, and the secondary outcomes were dentoalveolar changes, which were studied in both the conventional corticotomy and the flap-only procedure based on a cone-beam computed tomography (CBCT) wherein the alveolar bone density (BD) around canines, tipping, and rotational changes in canines, premolars, and molars were assessed.

Results

Corticotomy resulted in greater canine angulation, lesser canine rotation and premolar rotation, and greater molar rotation compared with flap elevation, but these differences were statistically insignificant.

Conclusion

Though the corticotomy resulted in higher BD, the differences were statistically insignificant. There was no significant difference in the rate of space closure assessed by the two techniques compared.

Antimicrobial Efficacy of Blended Essential Oil and Chlorhexidine against Periodontal Pathogen (P.gingivalis)-An In Vitro Study

Background

Essential oils (EOs) have a considerable amount of therapeutic and preventive effect in treating dental diseases due to their wider potential as antibacterial and anti-inflammatory agents. EOs like virgin coconut oil, eucalyptus oil, peppermint oil thyme oil, and clove oil, when used in combination, may further have enhanced antimicrobial effects. However, limited information exists on the synergistic effect of these oils when used in combination, especially on the primary periodontal pathogen Porphyromonas gingivalis.

Aim

The current study aims to compare the antimicrobial efficacy of commercially available EO on the periodontal pathogen, P. gingivalis, in comparison to chlorhexidine (CHX).

Materials and Methods

Antimicrobial efficacy of EO and CHX was assessed at various concentrations against the periodontal pathogen P. gingivalis, by evaluating the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC).

Results

P. gingivalis was seen to be sensitive at a MIC of 100 μg/ml and 50 μg/ml concentration of the EO, which is regarded as the MIC of EO against P. gingivalis and CHX effectively inhibited microbial growth at 0.4 μg/ml.

Conclusion

A combination of EOs possesses a potent antibacterial activity against P. gingivalis, and the antibacterial efficacy increases with increasing concentration of EOs.

Paternal stress alters synaptic density and expression of GAP-43, GRIN1, M1 and SYP genes in the hippocampus and cortex of offspring of stress-induced male rats

Adverse experiences during pregnancy have a negative impact on the neuronal structure and behavior of offspring, but the effects of a father's life events on the outcome of progeny are scarce. The present study is intended to investigate whether paternal stress affects the offspring brain structure, especially those regions concerned with learning and formation of memory, namely the hippocampus (HC) and prefrontal cortex (PFC), and also the expression of certain genes linked to learning and memory in the offspring. Induced stress to male rats by five stressors, one per day followed by allowing them to mate with the normal, unstressed female. Synaptophysin immunoreactivity was assessed in the tissue sections of the HC and PFC as well as expression of genes concerned with learning and memory was evaluated by RT-PCR in the progeny of stress-received males. The progeny of stressed rats had reduced antisynaptophysin immunoreactivity in the HC and PFC. The synaptic density in HC was less in the A-S (Offspring of male rats who received stress during adulthood) and PA-S (offspring of male rats who received stress during both adolescence and adulthood) than in P-S (offspring of male rats who received stress during adolescence) and C-C (offspring of control) groups. Similar results were observed even in the PFC. The results of post hoc tests proved that the HC and PFC of the progeny of stress-exposed rats exhibited considerably less synaptic density than control (P<0.05), and the levels of expression of GAP-43, GRIN1, M1, and SYP genes in HC and PFC were down-regulated. This study concludes that paternal adverse experiences can affect the offspring's synaptic plasticity and also the genes, which can regulate learning and formation of memory.

Enhancing the Breast Histopathology Image Analysis for Cancer Detection Using Variational Autoencoder

A breast tissue biopsy is performed to identify the nature of a tumour, as it can be either cancerous or benign. The first implementations involved the use of machine learning algorithms. Random Forest and Support Vector Machine (SVM) were used to classify the input histopathological images into whether they were cancerous or non-cancerous. The implementations continued to provide promising results, and then Artificial Neural Networks (ANNs) were applied for this purpose. We propose an approach for reconstructing the images using a Variational Autoencoder (VAE) and the Denoising Variational Autoencoder (DVAE) and then use a Convolutional Neural Network (CNN) model. Afterwards, we predicted whether the input image was cancerous or non-cancerous. Our implementation provides predictions with 73% accuracy, which is greater than the results produced by our custom-built CNN on our dataset. The proposed architecture will prove to be a new field of research and a new area to be explored in the field of computer vision using CNN and Generative Modelling since it incorporates reconstructions of the original input images and provides predictions on them thereafter.

Evaluation of cardiovascular risk scores after implementation of digital therapeutic intervention in patients with uncontrolled primary hypertension

Funding Acknowledgements

Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Terrals Technologies Private Limited

Background

Cardiovascular risk scores serve as a useful tool to reinforce the importance of lifestyle modifications and treatment compliance. Yet, there remains a gap between advice by the physician and its implementation by the patient. Digital therapeutics (DTx) as a technology based approach can help bridge this gap. In this context, knowing the quantum of change in disease outcomes and long term cardiovascular risk after DTx implementation can be helpful in persuading the patients for lifestyle interventions.

Purpose

To evaluate the change in cardiovascular risk scores in hypertensive patients after a digital therapeutic intervention.

Method

We conducted a prospective, single arm, 12-week intervention trial at two primary care sites in India. A total of 125 subjects within the age group of 30-65 years with primary hypertension (≥ 140/90 mmHg) were enrolled. They were provided DTx intervention which consisted of modification in diet, physical activities, self-monitoring and health education superimposed on behavior science theories. Atherosclerotic Cardiovascular Disease (ASCVD) risk scores from Pooled Cohort Equation were calculated for all subjects before & after the DTx intervention and the change in score was assessed.

Results

We received the data of 116 patients having a mean age of 47.66 ± 9.47 years and mean BMI of 27.20 ± 4.77 Kg/m². After DTx intervention, the change in mean systolic blood pressure, among other ASCVD equation parameters, was -26.38 mmHg (154.77 vs 128.39 mmHg, P&lt;0.001). The corresponding post-intervention 10-year mean ASCVD risk score for the sample population decreased by 4.74% (11.86% vs 7.12%, P&lt;0.001). The 10-year mean relative risk of ASCVD decreased by 39.95 %. Similarly ASCVD lifetime risk score decreased by 3.04% (49.54% vs 46.50%, P&lt;0.001), amounting to mean relative risk reduction of 6%. Female subjects (n=60) showed higher 10-year mean ASCVD risk reduction of 5.70% (11.70% vs 6.00%, P&lt;0.001), compared to male subjects (n=56) who showed 10-year mean ASCVD risk reduction of 3.76% (12.06% vs 8.30%, P&lt;0.001). This amounted to a relative risk reduction of 48.69% and 30.71% for female & male subjects, respectively. Patients with stage 1 hypertension (n=82) showed a reduction in 10-year mean ASCVD risk of 2.86% (9.90% vs 7.04%, P&lt;0.001). Patients with stage 2 (n=30) & stage 3 hypertension (n=4) achieved higher 10-year mean ASCVD risk reduction of 8.67% (16.07% vs 7.40%, P&lt;0.001) & 13.6% (20.88% vs 7.20%, P=.006).

Conclusion

Digital therapeutics enabled intervention has caused a significant decrease in 10-year & lifetime ASCVD risks which can eventually lead to a decrease in cardiovascular complications, morbidity and mortality. Therefore, such intervention programs should be further explored and studied.

Prospective evaluation of digital therapeutic intervention on blood pressure control in Indian patients with uncontrolled primary hypertension

Funding Acknowledgements

Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Terrals Technologies Private Limited

Background

Digital therapeutics (DTx) has emerged as a new approach in recent years making use of connected devices, smartphone software and electronic communication tools to help manage chronic cardiovascular diseases such as hypertension. Considering its nascent stage, many questions still remain with respect to its applicability, effectiveness and limitations.

Purpose

Assess the effectiveness of digital therapeutic intervention in controlling blood pressure in Indian patients with primary hypertension

Methods

We conducted a prospective, 12 week, single arm, interventional study, including 125 subjects at multiple sites in India. Subjects with uncontrolled primary hypertension (≥140 mmHg systolic and/or ≥90 mmHg diastolic) in the age group of 30-65 years who were under routine antihypertensive treatment were enrolled. They were provided supplementary DTx intervention comprising of smartphone application based personalized management of diet, exercise, self monitoring and health education by nutritionists and health coaches. Efficacy endpoints such as blood pressure (BP), body mass index (BMI), lipids, etc. were measured before and after the intervention.

Results

A total of 122 subjects who completed the trial had mean age of 47.44 ± 9.43 years, comprising 58 (48%) male and 64 (52%) female patients and mean BMI of 27.21 ±4.7 kg/m². Post-intervention change in mean systolic BP (SBP) was -25.61 mmHg (95% CI 22.55 - 28.66, 154.98 mmHg vs 129.37 mmHg, P&lt;.001). The post-intervention change in mean diastolic BP (DBP) was -18.62 mmHg (95% CI 16.29 - 20.96, 100.04 mmHg vs 81.42 mmHg, P&lt;.001). Patients with stage 1 hypertension (69.6%) achieved mean systolic reduction of 20.87 ± 13.9 mmHg, (148.54 mmHg vs 127.67 mmHg, P&lt;.001). The patients with stage 2 (27.04%) and stage 3 (3.27%) hypertension achieved a higher mean SBP reduction of 34.67 ± 15.19 mmHg (167.24 mmHg vs 132.58 mmHg, P&lt;.001) and 51.5 ± 40.45 mmHg (190.75 mmHg vs 139.25 mmHg, P=.084), respectively. Male and female patients showed almost similar reduction in systolic BP of 25.47 ± 14.72 and 25.73 ± 19.39 mmHg, respectively. Also, the subjects with other chronic comorbidities (23.9%) achieved a mean systolic BP reduction of 29.51 ± 15.25 mmHg.

A total of 79.51% (n=97) patients achieved the SBP &lt;140 mmHg, while 59.84% (n=73) patients achieved SBP &lt;130 mmHg. Combining both the systolic and diastolic targets together, 65.57% (n=80) patients achieved the target of &lt;140/90 mmHg, while ESC/ESH 2018 recommended BP target of &lt;130/80 mmHg was achieved by 31.15% (n=38) patients.

Conclusion

The implementation of Digital therapeutic application integrated with primary healthcare resulted in significant reduction in blood pressure in participants with uncontrolled hypertension. Greater improvement was observed in participants with higher baseline blood pressure.

Kids' Emotion Recognition Using Various Deep-Learning Models with Explainable AI

Human ideas and sentiments are mirrored in facial expressions. They give the spectator a plethora of social cues, such as the viewer's focus of attention, intention, motivation, and mood, which can help develop better interactive solutions in online platforms. This could be helpful for children while teaching them, which could help in cultivating a better interactive connect between teachers and students, since there is an increasing trend toward the online education platform due to the COVID-19 pandemic. To solve this, the authors proposed kids' emotion recognition based on visual cues in this research with a justified reasoning model of explainable AI. The authors used two datasets to work on this problem; the first is the LIRIS Children Spontaneous Facial Expression Video Database, and the second is an author-created novel dataset of emotions displayed by children aged 7 to 10. The authors identified that the LIRIS dataset has achieved only 75% accuracy, and no study has worked further on this dataset in which the authors have achieved the highest accuracy of 89.31% and, in the authors' dataset, an accuracy of 90.98%. The authors also realized that the face construction of children and adults is different, and the way children show emotions is very different and does not always follow the same way of facial expression for a specific emotion as compared with adults. Hence, the authors used 3D 468 landmark points and created two separate versions of the dataset from the original selected datasets, which are LIRIS-Mesh and Authors-Mesh. In total, all four types of datasets were used, namely LIRIS, the authors' dataset, LIRIS-Mesh, and Authors-Mesh, and a comparative analysis was performed by using seven different CNN models. The authors not only compared all dataset types used on different CNN models but also explained for every type of CNN used on every specific dataset type how test images are perceived by the deep-learning models by using explainable artificial intelligence (XAI), which helps in localizing features contributing to particular emotions. The authors used three methods of XAI, namely Grad-CAM, Grad-CAM++, and SoftGrad, which help users further establish the appropriate reason for emotion detection by knowing the contribution of its features in it.

Anti-microbial efficacy of root canal preparation in deciduous teeth with manual and rotary files: A randomized clinical trial

BACKGROUND

In a pulpectomy, the eradication of microbes from the primary root canal is accomplished through biomechanical preparation, which could be carried out with either manual or rotary instruments.

AIMS

The objective of this clinical trial was to evaluate the efficiency of manual K-files, H-files, and Kedo-S Square rotary files in reducing microbial flora after canal preparation in primary molars.

MATERIALS AND METHODS

This randomized clinical trial consisted of 45 primary molars requiring pulpectomy. The teeth were randomly allocated to one of the three groups: Group I: Manual K-files, Group II: Manual H-files, and Group III: Kedo-S Square rotary files, based on the type of instrumentation. Pre-and Post-instrumentation sampling was performed using clean absorbent paper points and kept in a clean Eppendorf tube having thioglycolate broth as the transport medium. Culturing was performed on agar media from which both aerobic and anaerobic microbial counts were estimated. Collected data were statistically analyzed using one-way analysis of variance (ANOVA) and Wilcoxon signed-rank test. Following root canal preparation, 87-89% reduction of the aerobic and anaerobic microbial load was noted in group I, whereas it was an 89-92% reduction in group II and a 93-95% reduction in group III.

RESULTS

Biomechanical preparation with Kedo-S Square rotary file showed higher efficacy in microbial reduction compared to manual instrumentation.

CONCLUSION

Manual and rotary files were equally effective in removing root canal microbes. Biomechanical preparation with a KedoS Square rotary file resulted in greater microbial efficacy. Hence In children, effective root canal cleaning in a short period of time is a major consideration.

Explainable AI for Bearing Fault Prognosis Using Deep Learning Techniques

Predicting bearing failures is a vital component of machine health monitoring since bearings are essential parts of rotary machines, particularly large motor machines. In addition, determining the degree of bearing degeneration will aid firms in scheduling maintenance. Maintenance engineers may be gradually supplanted by an automated detection technique in identifying motor issues as improvements in the extraction of useful information from vibration signals are made. State-of-the-art deep learning approaches, in particular, have made a considerable contribution to automatic defect identification. Under variable shaft speed, this research presents a novel approach for identifying bearing defects and their amount of degradation. In the proposed approach, vibration signals are represented by spectrograms, and deep learning methods are applied via pre-processing with the short-time Fourier transform (STFT). A convolutional neural network (CNN), VGG16, is then used to extract features and classify health status. After this, RUL prediction is carried out with the use of regression. Explainable AI using LIME was used to identify the part of the image used by the CNN algorithm to give the output. Our proposed method was able to achieve very high accuracy and robustness for bearing faults, according to numerous experiments.

Data augmentation using Variational Autoencoders for improvement of respiratory disease classification

Computerized auscultation of lung sounds is gaining importance today with the availability of lung sounds and its potential in overcoming the limitations of traditional diagnosis methods for respiratory diseases. The publicly available ICBHI respiratory sounds database is severely imbalanced, making it difficult for a deep learning model to generalize and provide reliable results. This work aims to synthesize respiratory sounds of various categories using variants of Variational Autoencoders like Multilayer Perceptron VAE (MLP-VAE), Convolutional VAE (CVAE) Conditional VAE and compare the influence of augmenting the imbalanced dataset on the performance of various lung sound classification models. We evaluated the quality of the synthetic respiratory sounds’ quality using metrics such as Fréchet Audio Distance (FAD), Cross-Correlation and Mel Cepstral Distortion. Our results showed that MLP-VAE achieved an average FAD of 12.42 over all classes, whereas Convolutional VAE and Conditional CVAE achieved an average FAD of 11.58 and 11.64 for all classes, respectively. A significant improvement in the classification performance metrics was observed upon augmenting the imbalanced dataset for certain minority classes and marginal improvement for the other classes. Hence, our work shows that deep learning-based lung sound classification models are not only a promising solution over traditional methods but can also achieve a significant performance boost upon augmenting an imbalanced training set.

Target-DBPPred: An intelligent model for prediction of DNA-binding proteins using discrete wavelet transform based compression and light eXtreme gradient boosting

DNA-protein interaction is a critical biological process that performs influential activities, including DNA transcription and recombination. DBPs (DNA-binding proteins) are closely associated with different kinds of human diseases (asthma, cancer, and AIDS), while some of the DBPs are used in the production of antibiotics, steroids, and anti-inflammatories. Several methods have been reported for the prediction of DBPs. However, a more intelligent method is still highly desirable for the accurate prediction of DBPs. This study presents an intelligent computational method, Target-DBPPred, to improve DBPs prediction. Important features from primary protein sequences are investigated via a novel feature descriptor, called EDF-PSSM-DWT (Evolutionary difference formula position-specific scoring matrix-discrete wavelet transform) and several other multi-evolutionary methods, including F-PSSM (Filtered position-specific scoring matrix), EDF-PSSM (Evolutionary difference formula position-specific scoring matrix), PSSM-DPC (Position-specific scoring matrix-dipeptide composition), and Lead-BiPSSM (Lead-bigram-position specific scoring matrix) to encapsulate diverse multivariate features. The best feature set from the features of each descriptor is selected using sequential forward selection (SFS). Further, four models are trained using Adaboost, XGB (eXtreme gradient boosting), ERT (extremely randomized trees), and LiXGB (Light eXtreme gradient boosting) classifiers. LiXGB, with the best feature set of EDF-PSSM-DWT, has attained 6.69% and 15.07% higher performance in terms of accuracies using training and testing datasets, respectively. The obtained results verify the improved performance of our proposed predictor over the existing predictors.

Improving the Accuracy of Ensemble Machine Learning Classification Models Using a Novel Bit-Fusion Algorithm for Healthcare AI Systems

Healthcare AI systems exclusively employ classification models for disease detection. However, with the recent research advances into this arena, it has been observed that single classification models have achieved limited accuracy in some cases. Employing fusion of multiple classifiers outputs into a single classification framework has been instrumental in achieving greater accuracy and performing automated big data analysis. The article proposes a bit fusion ensemble algorithm that minimizes the classification error rate and has been tested on various datasets. Five diversified base classifiers k- nearest neighbor (KNN), Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), Decision Tree (D.T.), and Naïve Bayesian Classifier (N.B.), are used in the implementation model. Bit fusion algorithm works on the individual input from the classifiers. Decision vectors of the base classifier are weighted transformed into binary bits by comparing with high-reliability threshold parameters. The output of each base classifier is considered as soft class vectors (CV). These vectors are weighted, transformed and compared with a high threshold value of initialized δ = 0.9 for reliability. Binary patterns are extracted, and the model is trained and tested again. The standard fusion approach and proposed bit fusion algorithm have been compared by average error rate. The error rate of the Bit-fusion algorithm has been observed with the values 5.97, 12.6, 4.64, 0, 0, 27.28 for Leukemia, Breast cancer, Lung Cancer, Hepatitis, Lymphoma, Embryonal Tumors, respectively. The model is trained and tested over datasets from UCI, UEA, and UCR repositories as well which also have shown reduction in the error rates.

Interventional Challenges in Non-Tubal Ectopic Pregnancy

Objective: Non-tubal ectopic pregnancies (EPs) are rare and potentially life threatening. The number is rising due to various risk factors and there are no uniform guidelines in the management of EPs. This study was done to assess risk factors and challenges in the management of EPs.

Materials and methods: This is a retrospective observational descriptive study that was done at SDM College of Medical Sciences & Hospital, Shri Dharmasthala Manjunatheshwara University Dharwad, Karnataka India. Data was collected from the medical records section of all the patients of non-tubal ectopic pregnancies managed in our hospital from January 2020 to June 2021. The collected data were analyzed for demographic characteristics, risk factors and management.

Results: The incidence of ectopic pregnancies in our institute was 6-7 per 1000 pregnancies, of which 20% of the ectopic pregnancies were non-tubal. The incidence was higher than the other studies, which could be due to our center being a tertiary care referral center. Cesarean scar ectopic pregnancies were the most common accounting for 60% of cases. The management varied from conservative to minimally invasive surgery to hysterectomy hysterectomy with bilateral internal iliac artery ligation, depending upon the clinical presentation, duration of gestation, presence of fetal cardiac activity and hemodynamic stability. The other non-tubal ectopic pregnancies were cervical, ovarian, corneal and heterotopic. Cervical pregnancy beyond 12 weeks of gestation was rare which was managed by conserving the uterus.

Conclusion: Non-tubal ectopic pregnancies are rare. Early diagnosis requires a high index of suspicion if missed can lead to an array of complications leading to loss of fertility, morbidity, and mortality. The key step to avert the complications is early diagnosis and individualized treatment.