Enhancing Skin Cancer Classification using Efficient Net B0-B7 through Convolutional Neural Networks and Transfer Learning with Patient-Specific Data.

BACKGROUND: Skin cancer diagnosis challenges dermatologists due to its complex visual variations across diagnostic categories. Convolutional neural networks (CNNs), specifically the Efficient Net B0-B7 series, have shown superiority in multiclass skin cancer classification. This study addresses the limitations of visual examination by presenting a tailored preprocessing pipeline designed for Efficient Net models. Leveraging transfer learning with pre-trained ImageNet weights, the research aims to enhance diagnostic accuracy in an imbalanced multiclass classification context. METHODS: The study develops a specialized image preprocessing pipeline involving image scaling, dataset augmentation, and artifact removal tailored to the nuances of Efficient Net models. Using the Efficient Net B0-B7 dataset, transfer learning fine-tunes CNNs with pre-trained ImageNet weights. Rigorous evaluation employs key metrics like Precision, Recall, Accuracy, F1 Score, and Confusion Matrices to assess the impact of transfer learning and fine-tuning on each Efficient Net variant's performance in classifying diverse skin cancer categories. RESULTS: The research showcases the effectiveness of the tailored preprocessing pipeline for Efficient Net models. Transfer learning and fine-tuning significantly enhance the models' ability to discern diverse skin cancer categories. The evaluation of eight Efficient Net models (B0-B7) for skin cancer classification reveals distinct performance patterns across various cancer classes. While the majority class, Benign Kertosis, achieves high accuracy (>87%), challenges arise in accurately classifying Eczema classes. Melanoma, despite its minority representation (2.42% of images), attains an average accuracy of 80.51% across all models. However, suboptimal performance is observed in predicting warts molluscum (90.7%) and psoriasis (84.2%) instances, highlighting the need for targeted improvements in accurately identifying specific skin cancer types. CONCLUSION: The study on skin cancer classification utilizes EfficientNets B0-B7 with transfer learning from ImageNet weights. The pinnacle performance is observed with EfficientNet-B7, achieving a groundbreaking top-1 accuracy of 84.4% and top-5 accuracy of 97.1%. Remarkably efficient, it is 8.4 times smaller than the leading CNN. Detailed per-class classification exactitudes through Confusion Matrices affirm its proficiency, signaling the potential of EfficientNets for precise dermatological image analysis.

Neural harmony: revolutionizing thyroid nodule diagnosis with hybrid networks and genetic algorithms.

In the contemporary world, thyroid disease poses a prevalent health issue, particularly affecting women's well-being. Recognizing the significance of maternal thyroid (MT) hormones in fetal neurodevelopment during the first half of pregnancy, this study introduces the HNN-GSO model. This groundbreaking hybrid approach, utilizing the MT dataset, integrates ResNet-50 and Artificial Neural Network (ANN) within a Glow-worm Swarm Optimization (GSO) framework for optimal parameter tuning. With a comprehensive methodology involving dataset preprocessing and Genetic Algorithm (GA) for feature selection, our model leverages ResNet-50 for feature extraction and ANN for classification tasks. Implemented in Python, the HNN-GSO model outperforms existing models, including K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Logistic Regression (LR), Random Forest (RF), ResNet, GoogleNet, and ANN, achieving an impressive accuracy rate of 98%. This success underscores the effectiveness of our approach in complex classification tasks within machine learning (ML) and pattern recognition, specifically tailored to the Thyroid Ultrasound Images (TUI) Dataset. To provide a comprehensive understanding of performance, additional statistical measures such as precision, recall, and F1 score were considered. The HNN-GSO model consistently outperformed competitors across these metrics, showcasing its superiority in MT classification. The HNN-GSO model seamlessly combines ResNet-50's feature extraction, ANN's classification robustness, and GSO's optimization for unparalleled performance. This research offers a promising framework for advancing ML methodologies, enhancing accuracy, and efficiency in classification tasks related to MT health.

Microalgal biorefineries: Advancement in machine learning tools for sustainable biofuel production and value-added products recovery.

The microalgae can be converted into biofuels, biochemicals, and bioactive compounds in a biorefinery. Recently, designing and executing more viable and sustainable biofuel production from microalgal biomass is one of the vital challenges in the development of biorefinery. Scalable cultivation of microalgae is mandatory for commercializing and industrializing the biorefinery. The intrinsic complication in cultivation of microalgae is the physiological and operational factors that renders challenging impact to enable a smooth and profitable operation. However, this aim can only be successful via a simulation prospect. Machine learning tools provides advanced approaches for evaluating, predicting, and controlling uncertainties in microalgal biorefinery for sustainable biofuel production. The present review provides a critical evaluation of the most progressing machine learning tools that validate a potential to be employed in microalgal biorefinery. These tools are highly potential for their extensive evaluation on microalgal screening and classification. However, the application of these tools for optimization of microalgal biomass cultivation in industries in order to increase the biomass production, is still in its initial stages. Integrated hybrid machine learning tools can aid the industries to function efficiently with least resources. Some of the challenges, and perspectives of machine learning tools are discussed. Besides, future prospects are also emphasized. Though, most of the research reports on machine learning tools are not appropriate to gather generalized information, standard protocols and strategies must be developed to design generalized machine learning tools. On a whole, this review offers a perspective information about digitalized microalgal exploitation in a microalgal biorefinery.

Biodegradation of microplastics: Advancement in the strategic approaches towards prevention of its accumulation and harmful effects.

Microplastics (MPs) are plastic particles in a size ranging from 1 mm to 5 mm in diameter, and are formed by the breakdown of plastics from different sources. They are emerging environmental pollutants, and pose a great threat to living organisms. Improper disposal, inadequate recycling, and excessive use of plastic led to the accumulation of MP in the environment. The degradation of MP can be done either biotically or abiotically. In view of that, this article discusses the molecular mechanisms that involve bacteria, fungi, and enzymes to degrade the MP polymers as the primary objective. As per as abiotic degradation is concerned, two different modes of MP degradation were discussed in order to justify the effectiveness of biotic degradation. Finally, this review is concluded with the challenges and future perspectives of MP biodegradation based on the existing research gaps. The main objective of this article is to provide the readers with clear insight, and ideas about the recent advancements in MP biodegradation.

Lignocellulosic biomass-based pyrolysis: A comprehensive review.

The efficacious application of lignocellulosic biomass for the new valuable chemicals generation curbs the excessive dependency on fossil fuels. Among the various techniques available, pyrolysis has garnered much attention for conversion of lignocellulosic biomass (encompasses cellulose, hemicellulose and lignin components) into product of solid, liquid and gases by thermal decomposition in an efficient manner. Pyrolysis conversion mechanism can be outlined as formation of char, depolymerisation, fragmentation and other secondary reactions. This paper gives a deep insight about the pyrolytic behavior of the lignocellulosic components accompanied by its by-products. Also several parameters such as reaction environment, temperature, residence time and heating rate which has a great impact on the pyrolysis process are also elucidated in a detailed manner. In addition the environmental and economical facet of lignocellulosic biomass pyrolysis for commercialization at industrial scale is critically analyzed. This article also illustrates the prevailing challenges and inhibition in implementing lignocellulosic biomass based pyrolysis with possible solution.

Efficacy of a Novel Topical Combination of Fipronil 9.8% and (S)-Methoprene 8.8% against Ticks and Fleas in Naturally Infested Dogs.

The efficacy of a novel topical combination of fipronil 9.8% (w/v) and (S)-methoprene 8.8% (w/v) (Fiprofort® Plus) was tested against ticks and fleas in naturally infested dogs. A total of fifty dogs were allocated in the study with ticks infestation (n = 35) and fleas infestation (n = 15). On day 0, thirty-five tick and fifteen flea infested dogs received the test formulation, a combination of fipronil 9.8% (w/v) and (S)-methoprene 8.8% (w/v) spot-on solution. Ticks and flea counts were taken on days 0 (pretreatment) and 3, 7, 14, 21, 28, and 35 after treatment. Blood samples were collected for evaluation of haematological parameters on days 0 (pretreatment) and 7, 21, and 35 after treatment. All the adult ticks and fleas collected were identified as Rhipicephalus sanguineus and Ctenocephalides felis, respectively. The efficacy of spot-on formulation against ticks was 34.00% (day 3), 53.14% (day 7), 62.71% (day 14), 65.48% (day 21), 59.80% (day 28), and 58.82% (day 35), whereas against fleas it was 38.00% (day 3), 64.34% (day 7), 89.67% (day 14), 95.40% (day 21), 100.00% (day 28), and 100.00% (day 35). Haematological parameters for ticks and fleas infested dogs were statistically nonsignificant as compared to control. The combination of fipronil and (S)-methoprene eliminated the existing ticks and fleas infestation and prevented the dogs from flea and tick infestation for four weeks.

Modified pelvic floor repair in prolapsed uterus, with total prevention of vault prolapse.

In and around Ammapettai, a rural area with an economically backward community, 515 cases were operated for prolapsed uteri, by doing modified pelvic floor repairs (Dr. Sunthanthradevi's method), without any incidence of vault prolapse, with patients being followed up for two years after their surgeries.

Corticotomy-assisted retraction: an outcome assessment.

BACKGROUND: To assess the efficiency and treatment outcome of patients treated with corticotomy-assisted en-masse orthodontic retraction as compared with the en-masse retraction without corticotomy. MATERIALS AND METHODS: Forty adult patients with bimaxillary protrusion requiring correction of bidental proclination constituted the sample. The study group consisted of 22 patients (male 11, female 11) willing to undergo surgery to reduce the duration of their orthodontic treatment and 18 patients (male 9, female 9) desirous of undergoing conventional orthodontic treatment without surgical intervention constituted the control group. Comparison of rate of retraction and anchor loss between the study and the control group was assessed. RESULTS: Average rate of space closure of 1.8 mm/month in the maxilla and 1.57 mm/month in the mandible was observed in the study group compared to 1.02 mm/month in the maxilla and 0.87 mm/month in the mandible in the control group. The rate of retraction accelerated during the first 2 months of retraction. Molar anchor loss of approximately 0.6 mm occurred in the study group, and 1.8 mm occurred in the control group during the 4 months. CONCLUSION: The rate of retraction with study group was twice as faster when compared to the control group, accelerating during the first 2 months of retraction. There was better anchorage control with the undecorticated molar segment during the retraction period but was found to increase as time advanced.

Modulation of voltage sensitivity by N-terminal cytoplasmic residues in human Kv1.2 channels.

Potassium channels are now among the best understood membrane proteins and most salient functions have been mapped onto distinct portions of the protein. The detailed mechanism by which movement of the voltage sensor is transduced into channel opening is yet to be understood. We have constructed chimaeras from our collection of human voltage-gated potassium channels and expressed them in Xenopus oocytes. Here we report on a chimaeric construct, 1N/2, generated by swapping the N-terminal cytoplasmic residues of hKv1.1 onto the transmembrane body of hKv1.2. This chimaera functions as a classic outward rectifier but with a 25 mV hyperpolarizing shift in the mid-point of channel activation. The conductance of oocytes expressing this construct decreases significantly on depolarizing beyond +5 mV, unlike full-length hKv1.2. Other parameters such as ionic selectivity and charybdotoxin blockage are unaffected in making the chimaera. These observations suggest that the introduction of the "foreign" chain from hKv1.1 does not cause a large-scale perturbation of channel structure. Loss of the N-terminus from hKv1.2 is not responsible for the shift in voltage dependence, as a truncation construct, delta75N2, starting at the splice junction, has the same voltage-dependence as full-length hKv1.2. Both constructs show a maximum in their conductance-voltage curves. This decline in conductance on extensive depolarization may arise due to perturbations to the machinery that locks channels into their open state on depolarization. Taken together with our observations on other N-terminal swapped chimaeras, our data imply that N-terminal residues can interact with transmembrane regions and perturb the machinery mediating voltage-dependent channel gating.