AI-driven estimation of O6 methylguanine-DNA-methyltransferase (MGMT) promoter methylation in glioblastoma patients: a systematic review with bias analysis

BACKGROUND

Accurate and non-invasive estimation of MGMT promoter methylation status in glioblastoma (GBM) patients is of paramount clinical importance, as it is a predictive biomarker associated with improved overall survival (OS). In response to the clinical need, recent studies have focused on the development of non-invasive artificial intelligence (AI)-based methods for MGMT estimation. In this systematic review, we not only delve into the technical aspects of these AI-driven MGMT estimation methods but also emphasize their profound clinical implications. Specifically, we explore the potential impact of accurate non-invasive MGMT estimation on GBM patient care and treatment decisions.

METHODS

Employing a PRISMA search strategy, we identified 33 relevant studies from reputable databases, including PubMed, ScienceDirect, Google Scholar, and IEEE Explore. These studies were comprehensively assessed using 21 diverse attributes, encompassing factors such as types of imaging modalities, machine learning (ML) methods, and cohort sizes, with clear rationales for attribute scoring. Subsequently, we ranked these studies and established a cutoff value to categorize them into low-bias and high-bias groups.

RESULTS

By analyzing the 'cumulative plot of mean score' and the 'frequency plot curve' of the studies, we determined a cutoff value of 6.00. A higher mean score indicated a lower risk of bias, with studies scoring above the cutoff mark categorized as low-bias (73%), while 27% fell into the high-bias category.

CONCLUSION

Our findings underscore the immense potential of AI-based machine learning (ML) and deep learning (DL) methods in non-invasively determining MGMT promoter methylation status. Importantly, the clinical significance of these AI-driven advancements lies in their capacity to transform GBM patient care by providing accurate and timely information for treatment decisions. However, the translation of these technical advancements into clinical practice presents challenges, including the need for large multi-institutional cohorts and the integration of diverse data types. Addressing these challenges will be critical in realizing the full potential of AI in improving the reliability and accessibility of MGMT estimation while lowering the risk of bias in clinical decision-making.

GC/MS Profiling and Evaluation of Leaf Essential Oil for Bactericidal Effect and Free Radical Scavenging Activity of Plectranthus amboinicus (Lour.) Spreng Collected from Odisha, India

Plectranthus amboinicus (Lour.) Spreng, known as the Indian borage or Mexican mint, is one of the most documented species in the family Lamiaceae for its therapeutic and pharmaceutical values. It is found in the tropical and subtropical regions of the world. The leaf essential oil has immense medicinal benefits like treating illnesses of the skin and disorders like colds, asthma, constipation, headaches, coughs, and fevers. After analyzing earlier reports with regard to the quantity and quality of leaf oil yield, we discovered that the germplasm taken from Odisha is preferable to other germplasms. The objective of the present work is to evaluate the free radical scavenging activity and bactericidal effect of leaf essential oil (EO) of Plectranthus amboinicus (Lour.) Spreng collected from the state of Odisha, India. The hydro distillation technique has been used for essential oil extraction. Upon GC/MS analysis, approximately 57 compounds were identified with Carvacrol as the major compound (peak area=20.25 %), followed by p-thymol (peak area=20.17 %), o-cymene (peak area=19.41 %) and carene (peak area=15.89 %). On evaluation of free radical scavenging activity, it was recorded that the best value of inhibitory concentration, was for DPPH with IC₅₀ =18.64 ppm and for H₂ O₂ with IC₅₀ =9.35 ppm. The EO showed efficient bactericidal effect against both gram positive (Mycobacterium smegmatis, Staphylococcus aureus, Enterococcus faecium) and gram negative (Escherichia coli, Vibrio cholerae, Klebsiella pneumoniae) bacteria studied through well diffusion method. Fumigatory action of the essential oil was found against M. smegmatis, the model organism for tuberculosis study. Alamar Blue assay, gave a result with MIC value for M. smegmatis i. e., 0.12 μg/ml and the MBC value of 0.12 μg/ml. Hence, P. amboinicus found in Odisha can be suggested as an elite variety and should be further investigated for efficient administration in drug formulation.

Fused deep learning paradigm for the prediction of o6-methylguanine-DNA methyltransferase genotype in glioblastoma patients: A neuro-oncological investigation

BACKGROUND

The O6-methylguanine-DNA methyltransferase (MGMT) is a deoxyribonucleic acid (DNA) repairing enzyme that has been established as an essential clinical brain tumor biomarker for Glioblastoma Multiforme (GBM). Knowing the status of MGMT methylation biomarkers using multi-parametric MRI (mp-MRI) helps neuro-oncologists to analyze GBM and its treatment plan.

METHOD

The hand-crafted radiomics feature extraction of GBM's subregions, such as edema(ED), tumor core (TC), and enhancing tumor (ET) in the machine learning (ML) framework, was investigated using support vector machine(SVM), K-Nearest Neighbours (KNN), random forest (RF), LightGBM, and extreme gradient boosting (XGB). For tissue-level analysis of the promotor genes in GBM, we used the deep residual neural network (ResNet-18) with 3D architecture, followed by EfficientNet-based investigation for variants as B0 and B1. Lastly, we analyzed the fused deep learning (FDL) framework that combines ML and DL frameworks.

RESULT

Structural mp-MRI consisting of T1, T2, FLAIR, and T1GD having a size of 400 and 185 patients, respectively, for discovery and replication cohorts. Using the CV protocol in the ResNet-3D framework, MGMT methylation status prediction in mp-MRI gave the AUC of 0.753 (p < 0.0001) and 0.72 (p < 0.0001) for the discovery and replication cohort, respectively. We presented that the FDL is ∼7% superior to solo DL and ∼15% to solo ML.

CONCLUSION

The proposed study aims to provide solutions for building an efficient predictive model of MGMT for GBM patients using deep radiomics features obtained from mp-MRI with the end-to-end ResNet-18 3D and FDL imaging signatures.

Effect of Learning Parameters on the Performance of the U-Net Architecture for Cell Nuclei Segmentation from Microscopic Cell Images

Nuclei segmentation of cells is the preliminary and essential step of pathological image analysis. However, robust and accurate cell nuclei segmentation is challenging due to the enormous variability of staining, cell sizes, morphologies, cell adhesion, or overlapping of the nucleus. The automation process to find the cell’s nuclei is a giant leap in this direction and have an important step toward bioimage analysis employing software tools. This article extensively analyzes deep U-Net architecture and has been applied to the Data Science Bowl dataset to segment the cell nuclei. The dataset undergoes various pre-processing tasks such as resizing, intensity normalization, and data augmentation prior to segmentation. The complete dataset then undergoes rigorous training and validation process to find the optimized hyperparameters and then the optimized model selection. The mean (m) ± standard deviation (sd) of Intersection over Union (IOU), and F1-Score (Dice score) have been calculated along with Accuracy during the training and validation process, respectively. The optimized U-net model results in training IOU of 0.94 ± 0.16 (m ± s), F1-score of 0.94 ± 0.17 (m ± s), training accuracy of 95.54, and validation accuracy of 95.45. With that model, we have applied a completely independent test cohort of the dataset and obtained the mean IOU of 0.93, F1-score of 0.9311, and mean Accuracy of 94.12, respectively.

Brain Tumor Characterization Using Radiogenomics in Artificial Intelligence Framework

Brain tumor characterization (BTC) is the process of knowing the underlying cause of brain tumors and their characteristics through various approaches such as tumor segmentation, classification, detection, and risk analysis. The substantial brain tumor characterization includes the identification of the molecular signature of various useful genomes whose alteration causes the brain tumor. The radiomics approach uses the radiological image for disease characterization by extracting quantitative radiomics features in the artificial intelligence (AI) environment. However, when considering a higher level of disease characteristics such as genetic information and mutation status, the combined study of "radiomics and genomics" has been considered under the umbrella of "radiogenomics". Furthermore, AI in a radiogenomics' environment offers benefits/advantages such as the finalized outcome of personalized treatment and individualized medicine. The proposed study summarizes the brain tumor's characterization in the prospect of an emerging field of research, i.e., radiomics and radiogenomics in an AI environment, with the help of statistical observation and risk-of-bias (RoB) analysis. The PRISMA search approach was used to find 121 relevant studies for the proposed review using IEEE, Google Scholar, PubMed, MDPI, and Scopus. Our findings indicate that both radiomics and radiogenomics have been successfully applied aggressively to several oncology applications with numerous advantages. Furthermore, under the AI paradigm, both the conventional and deep radiomics features have made an impact on the favorable outcomes of the radiogenomics approach of BTC. Furthermore, risk-of-bias (RoB) analysis offers a better understanding of the architectures with stronger benefits of AI by providing the bias involved in them.

Mechanisms associated with the rapid decline in sea ice cover around a stranded ship in the Lazarev Sea, Antarctica

In the satellite data era starting from 1979, the extent of Antarctic sea ice increased moderately for the first 37 years. However, the extent decreased to record low levels from 2016 to 2020, with the drop being greatest in the Weddell and Lazarev Seas of the Southern Ocean. An important question for the scientific fraternity and policymakers is to understand what ocean-atmospheric processes triggered such a rapid decline in sea ice. We employ in-situ, satellite, and atmospheric reanalysis data to examine the causative mechanism of anomalous sea ice variability in the Lazarev Sea at a time of ice growth in the annual cycle (March-April 2019), when a cargo ship was stuck in extensive ice cover and freed following the unusual decline in sea ice. High-resolution Sentinel-1 synthetic aperture radar captured a distinct view of the ship location and track within extensive ice cover of fast sea ice, dense pack ice, and icebergs in the Lazarev Sea on 27 March 2019. Subsequently, the sea ice cover declined and reached the fourth lowest extent in the entire satellite record during April 2019 which was 25.6% lower than the long-term mean value of 2.65 × 10⁶ km². We show that the anomalous sea ice variability was due to the occurrence of eastward-moving polar cyclones, including a quasi-stationary explosive development that impacted sea ice through extreme changes in ocean-atmospheric conditions. The cyclone-induced dynamic (poleward propagation of ocean waves and ice motion) and thermodynamic (heat and moisture plumes from midlatitudes, ocean mixed layer warming) processes coupled with high tides provided a conducive environment for an exceptional decline in sea ice over the region of ship movement.

Journey of MOSFET from Planar to Gate All Around: A Review

With the continuous miniaturization in device dimension to reach the expectation raised by semiconductor users, the shape and size of the MOSFET are changing periodically. The journey started in the year 1960, reached the milestone, and still going on to create history. Due to continuous downscaling, the device dimensions have already reached the critical limit and further miniaturization is a challenge. As a result of which some unwanted effects were raised unknowingly to suppress the device performances while entering into nanoscale. To overcome these kinds of barriers, different device architectures were proposed to keep the journey on. This paper focused on those types of advanced structures in MOSFET, which kept Moore's law alive.

Decadal changes in global phytoplankton compositions influenced by biogeochemical variables

The global environmental changes owing to natural and anthropogenic influences are challenging the structure and functioning of the ocean ecosystem. The complex processes interacting within the physical, chemical, and biological environment at different spatio-temporal scales and their impact on the ocean ecosystem processes are yet to be investigated. A long term trend on phytoplankton biomass in terms of Chlorophyll-a concentration (Chl-a), phytoplankton compositions and the processes that control the variability is required for understanding the ocean ecosystem. This study investigated decadal trends (2002-2015) of phytoplankton composition and biogeochemical parameters over the Global Ocean (GO), Southern Ocean (SO), and the Arctic Ocean (AO) using ocean color remote sensing and assimilated data from the National Aeronautics and Space Administration (NASA) Ocean biogeochemical model. The results revealed the dominance of larger cell phytoplankton mainly diatoms throughout the SO and AO; however, the coccolithophores dominate in the remaining part of the GO. Analysis of nutrients showed that nitrate is not a limiting factor for the variability of phytoplankton biomass in the SO and AO. The low nitrate concentration influenced in the rest of the GO. The photosynthetically available radiation (PAR) limiting the phytoplankton biomass and composition in the SO and AO. Although the SO is known as the high nutrient low chlorophyll (HNLC) region of the GO, the low iron concentration along with the PAR co-limits the growth of phytoplankton biomass. Trend analysis showed that an increase in Chl-a and diatoms in the SO and AO. In contrast, it declined significantly in the other regions of the GO, in response to the consistent increase in sea surface temperature. The results indicated that, shifting of phytoplankton community from regional to global scale have a greater implication for climate change and marine ecosystem.

Cinnamomum zeylanicum (Cinnamon) Derived Phytochemicals against Aspartate Semialdehyde Dehydrogenase of Aspergillus fumigatus Causing Aspergillosis

Aspergillosis is an infection caused by a type of fungus. The illnesses is caused by aspergillosis infection and usually attack the respiratory system, but their signs and severity vary greatly. It has already been established that plant Cinnamomum zeylanicum belonging to the Lauraceae family has the potential to help controlling aspergillosis. This work is focused on the identification of the particular phytochemical responsible for inhibiting and decreasing the effect of aspergillosis. Molecular docking method applied using “Biovia Discovery Studio”. “High positive values of -CDOCKER energy and -CDOCKER interaction energy” suggested that cinnamic acid can effectively deactivate aspergillosis.

The burden of diabetes and impaired glucose tolerance in India using the WHO 1999 criteria: prevalence of diabetes in India study (PODIS)

This random multistage cross-sectional population survey was undertaken to determine the prevalence of diabetes mellitus (DM) and impaired glucose tolerance (IGT) in subjects aged 25 years and above in India. The study was carried out in 77 centers (40 urban and 37 rural). 18363 (9008 males and 9355 females) subjects were studied. 10617 (5379 males and 5238 females) were from urban areas and 7746 (3629 males and 4117 females) from rural areas. Blood samples were taken after a fast of 10-12 h and 2 h after 75 g of oral glucose. Subjects were categorized as having IGT or DM using the World Health Organisation (WHO) (1999) criteria. The standardized prevalence rate for DM in the total Indian, urban and rural populations was 4.3, 5.9 and 2.7%, respectively. The corresponding IGT rates in the three populations was 5.2, 6.3 and 3.7%, respectively. The urban prevalence of DM and IGT was significantly greater than in the rural population (P < 0.001 in both instances). The prevalence of DM was significantly, more than that of IGT (P < 0.001) within both the rural and urban populations. Type 2 diabetes is a major health problem is India.

Comparing the ADA 1997 and the WHO 1999 criteria: Prevalence of Diabetes in India Study

This random multistage cross-sectional population survey was undertaken to determine the prevalence of type 2 diabetes mellitus (DM) in subjects aged 25 years and above in India. The study was carried out in 77 centres (42 urban and 35 rural) to reflect the size and heterogeneity of the Indian population. 18,363 (9008 male and 9355 female) subjects were studied. 10,617 (5379 males and 5238 females) were from urban areas and 7746 (3629 males and 4117 females) from rural areas. Blood samples were taken after a fast of 10-12 and 2 h after 75 g of oral glucose. Subjects were categorized as having impaired fasting glycemia (IFG) or DM using the 1997 ADA or having impaired glucose tolerance (IGT) or DM using the 1999 WHO criteria. The age- and gender-standardized prevalence rate for DM using the ADA criteria was 3.6% whilst that using the WHO criteria was 4.3% (P < 0.001). The respective standardized prevalence of DM, using the two criteria was, 4.7 and 5.6%, respectively (P < 0.001) in the urban Indian population and 2.0 and 2.7% (P < 0.02) in the rural Indian population. Using the WHO criteria, 581 subjects were newly diagnosed whilst the ADA criteria newly diagnosed 437 subjects. The respective numbers for the urban population were 425 and 323, and for the rural population were 146 and 114, respectively. The ADA criteria could diagnose 75.2, 76.0 and 73.0% of the subjects who had DM as per the WHO criteria. Of 739 Indian subjects who had IFG, 106 (14.3%) were diagnosed as having DM by the WHO criteria whilst 505 (68.3%) had values compatible with a diagnosis of IGT. Of the 536 urban subjects with IFG, 74 (13.8%) had DM and 350 (65.3%) had IGT using the WHO criteria. Of the 302 rural subjects with IFG, 32 (15.8%) had DM and 155 (76.3%) had IGT using the WHO criteria. 505 (49.9%) of 1012 Indian subjects with IGT as per the WHO criteria had IFG. 350 (47.7%) of 733 urban subjects and 155 (55.5%) of 279 rural subjects with IGT had values compatible with IFG as per the ADA criteria. Type 2 diabetes is a major health problem is India. The use of the ADA criteria would underestimate the prevalence of DM by not diagnosing subjects showing a poor response to a glucose challenge. This along with the discrepancies between subjects showing IGF or IGT could be a challenge to any prevention program.

The burden of diabetes and impaired fasting glucose in India using the ADA 1997 criteria: prevalence of diabetes in India study (PODIS)

This random multistage cross sectional population survey was undertaken to determine the prevalence of diabetes mellitus (DM) and impaired fasting glycemia/glucose (IFG) in subjects aged 25 years and above in India. The study was carried out in 108 centers (49 urban and 59 rural) to reflect the size and heterogeneity of the Indian population. 41,270 (20,534 males and 20,736 females) subjects were studied. 21,516 (10,865 males and 10,651 females) were from urban areas and 19,754 (9669 males and 10,085 females) from rural areas. Blood samples were taken after a fast of 10-12h and the subjects were categorized as having IFG or DM using the 1997 American Diabetes Association criteria. The age and gender standardized prevalence rate for DM and IFG in the total Indian population was 3.3 and 3.6% respectively (P < 0.001). The standardized prevalence of DM and IFG in urban areas was significantly higher than that for the rural population (urban DM prevalence 4.6% versus rural DM prevalence 1.9%, P < 0.001; urban IFG prevalence 4.8% versus rural IFG prevalence 2.5%, P < 0.001). There was no statistically significant difference in the prevalence between DM (4.6%) and IFG (4.8%) in the urban population. The rural prevalence of IFG (2.5%) was significantly (P <0.001) more than the rural prevalence of DM (1.9%). Type 2 diabetes is a major health problem is India.