Diagnostic accuracy of artificial intelligence assisted clinical imaging in the detection of oral potentially malignant disorders and oral cancer: A systematic review and meta-analysis

BACKGROUND

The objective of this study is to examine the application of AI algorithms in detecting OPMD and oral cancerous lesions, and to evaluate the accuracy variations among different imaging tools employed in these diagnostic processes.

MATERIALS AND METHODS

A systematic search was conducted in four databases: Embase, Web of Science, PubMed, and Scopus. The inclusion criteria included studies using machine learning algorithms to provide diagnostic information on specific oral lesions, prospective or retrospective design, and inclusion of OPMD. Sensitivity and specificity analyses were also required. Forest plots were generated to display overall diagnostic odds ratio (DOR), sensitivity, specificity, negative predictive values, and summary receiver operating characteristic (SROC) curves. Meta-regression analysis was conducted to examine potential differences among different imaging tools.

RESULTS

The overall DOR for AI-based screening of OPMD and oral mucosal cancerous lesions from normal mucosa was 68.438 (95%CI= [39.484, 118.623], I2 = 86%). The area under the SROC curve was 0.938, indicating excellent diagnostic performance. AI-assisted screening showed a sensitivity of 89.9% (95%CI= [0.866,0.925]; I2 = 81%), specificity of 89.2% (95%CI= [0.851,0.922], I2 = 79%), and a high negative predictive value of 89.5% (95%CI= [0.851; 0.927], I2 = 96%). Meta-regression analysis revealed no significant difference among the three image tools. After generating a GOSH plot, the DOR was calculated to be 49.30, and the area under the SROC curve was 0.877. Additionally, sensitivity, specificity, and negative predictive value were 90.5% (95%CI [0.873,0.929], I2=4%), 87.0% (95%CI [0.813,0.912], I2=49%) and 90.1% (95%CI [0.860,0.931], I2=57%), respectively. Subgroup analysis showed that clinical photography had the highest diagnostic accuracy.

CONCLUSIONS

AI-based detection using clinical photography shows a high diagnostic odds ratio and is easily accessible in the current era with billions of phone subscribers globally. This indicates that there is significant potential for AI to enhance the diagnostic capabilities of general practitioners to the level of specialists by utilizing clinical photographs, without the need for expensive specialized imaging equipment.

A single-cell atlas of conventional central chondrosarcoma reveals the role of endoplasmic reticulum stress in malignant transformation

The transformation of benign lesions to malignant tumours is a crucial aspect of understanding chondrosarcomas, which are malignant cartilage tumours that could develop from benign chondroid lesions. However, the process of malignant transformation for chondroid lesions remains poorly understood, and no reliable markers are available to aid clinical decision-making. To address this issue, we conducted a study analysing 11 primary cartilage tumours and controls using single-cell RNA sequencing. By creating a single-cell atlas, we were able to identify the role of endoplasmic reticulum (ER) stress in the malignant transformation of conventional central chondrosarcomas (CCCS). Our research revealed that lower levels of ER stress promote chondrosarcoma growth in a patient-derived xenograft mouse model, while intensive ER stress reduces primary chondrosarcoma cell viability. Furthermore, we discovered that the NF-κB pathway alleviates ER stress-induced apoptosis during chondrosarcoma progression. Our single-cell signatures and large public data support the use of key ER stress regulators, such as DNA Damage Inducible Transcript 3 (DDIT3; also known as CHOP), as malignant markers for overall patient survival. Ultimately, our study highlights the significant role that ER stress plays in the malignant transformation of cartilaginous tumours and provides a valuable resource for future diagnostic markers and therapeutic strategies.

Multi-Level Biomarkers for Early Diagnosis of Ischaemic Stroke: A Systematic Review and Meta-Analysis

Blood biomarkers hold potential for the early diagnosis of ischaemic stroke (IS). We aimed to evaluate the current weight of evidence and identify potential biomarkers and biological pathways for further investigation. We searched PubMed, EMBASE, the Cochrane Library and Web of Science, used R package meta4diag for diagnostic meta-analysis and applied Gene Ontology (GO) analysis to identify vital biological processes (BPs). Among 8544 studies, we included 182 articles with a total of 30,446 participants: 15675 IS, 2317 haemorrhagic stroke (HS), 1798 stroke mimics, 846 transient ischaemic attack and 9810 control subjects. There were 518 pooled biomarkers including 203 proteins, 114 genes, 108 metabolites and 88 transcripts. Our study generated two shortlists of biomarkers for future research: one with optimal diagnostic performance and another with low selection bias. Glial fibrillary acidic protein was eligible for diagnostic meta-analysis, with summary sensitivities and specificities for differentiating HS from IS between 3 h and 24 h after stroke onset ranging from 73% to 80% and 77% to 97%, respectively. GO analysis revealed the top five BPs associated with IS. This study provides a holistic view of early diagnostic biomarkers in IS. Two shortlists of biomarkers and five BPs warrant future investigation.

MOF-mediated acetylation of SIRT6 disrupts SIRT6-FOXA2 interaction and represses SIRT6 tumor-suppressive function by upregulating ZEB2 in NSCLC

Mammalian sirtuin 6 (SIRT6) regulates a spectrum of vital biological processes and has long been implicated in the progression of cancer. However, the mechanisms underlying the regulation of SIRT6 in tumorigenesis remain elusive. Here, we report that the tumor-suppressive function of SIRT6 in non-small cell lung cancer (NSCLC) is regulated by acetylation. Specifically, males absent on the first (MOF) acetylates SIRT6 at K128, K160, and K267, resulting in a decreased deacetylase activity of SIRT6 and attenuated SIRT6 tumor-suppressive function in NSCLC. Mechanistically, MOF-mediated SIRT6 acetylation hinders the interaction between SIRT6 and transcriptional factor FOXA2, which in turn leads to the transcriptional activation of ZEB2, thus promoting NSCLC progression. Collectively, these data indicate an acetylation-dependent mechanism that modulates SIRT6 tumor-suppressive function in NSCLC. Our findings suggest that the MOF-SIRT6-ZEB2 axis may represent a promising therapeutic target for the management of NSCLC.

Sex Differences in Association Between Gut Microbiome and Essential Hypertension Based on Ambulatory Blood Pressure Monitoring

BACKGROUND

Sex differences in the pathogenesis of hypertension exist. While gut microbiota (GM) has been associated with hypertension, it is unclear whether there are sex-linked differences in the association between GM and hypertension.

METHODS

We conducted a cross-sectional study to investigate the sex differences in associations between GM characterized by shotgun sequencing, GM-derived short-chain fatty acids, and 24-hour ambulatory blood pressure in 241 Hong Kong Chinese (113 men and 128 women; mean age, 54±6 years).

RESULTS

The hypertensive group was associated with GM alterations; however, significant differences in β-diversity and GM composition in hypertensive versus normotensive groups were only observed in women and not in men under various statistical models adjusting for the following covariates: age, sex, body mass index, sodium intake estimated by spot urine analysis, blood glucose, triglycerides, low- and high-density lipoprotein cholesterol, smoking, menopause, and fatty liver status. Specifically, Ruminococcus gnavus, Clostridium bolteae, and Bacteroides ovatus were significantly more abundant in the hypertensive women, whereas Dorea formicigenerans was more abundant in the normotensive female group. No bacterial species were found to be significantly associated with hypertension in men. Furthermore, total plasma short-chain fatty acids and propionic acid were independent predictors of systolic and diastolic blood pressure in women but not men.

CONCLUSONS

GM dysregulation was strongly associated with 24-hour ambulatory blood pressure in women but not men, which may be mediated through propionic acid. Our work suggests that sex differences may be an important consideration while assessing the role of GM in the development and treatment of hypertension.

Deficiency of exchange protein directly activated by cAMP (EPAC)-1 in mice augments glucose intolerance, inflammation, and gut dysbiosis associated with Western diet

BACKGROUND

Gut microbiota (GM) dysregulation, known as dysbiosis, has been proposed as a crucial driver of obesity associated with "Western" diet (WD) consumption. Gut dysbiosis is associated with increased gut permeability, inflammation, and insulin resistance. However, host metabolic pathways implicated in the pathophysiology of gut dysbiosis are still elusive. Exchange protein directly activated by cAMP (Epac) plays a critical role in cell-cell junction formation and insulin secretion. Here, we used homozygous Epac1-knockout (Epac1^-/-), Epac2-knockout (Epac2^-/-), and wild-type (WT) mice to investigate the role of Epac proteins in mediating gut dysbiosis, gut permeability, and inflammation after WD feeding.

RESULTS

The 16S rRNA gene sequencing of fecal DNA showed that the baseline GM of Epac2^-/-, but not Epac1^-/-, mice was represented by a significantly higher Firmicutes to Bacteroidetes ratio and significant alterations in several taxa compared to WT mice, suggesting that Epac2^-/- mice had gut dysbiosis under physiological conditions. However, an 8-week WD led to a similar gut microbiome imbalance in mice regardless of genotype. While Epac1 deficiency modestly exacerbated the WD-induced GM dysbiosis, the WD-fed Epac2^-/- mice had a more significant increase in gut permeability than corresponding WT mice. After WD feeding, Epac1^-/-, but not Epac2^-/-, mice had significantly higher mRNA levels of tumor necrosis factor-alpha (TNF-α) and F4/80 in the epididymal white adipose tissue (EWAT), increased circulating lipocalin-2 protein and more severe glucose intolerance, suggesting greater inflammation and insulin resistance in WD-fed Epac1^-/- mice than corresponding WT mice. Consistently, Epac1 protein expression was significantly reduced in the EWAT of WD-fed WT and Epac2^-/- mice.

CONCLUSION

Despite significantly dysregulated baseline GM and a more pronounced increase in gut permeability upon WD feeding, WD-fed Epac2^-/- mice did not exhibit more severe inflammation and glucose intolerance than corresponding WT mice. These findings suggest that the role of gut dysbiosis in mediating WD-associated obesity may be context-dependent. On the contrary, we demonstrate that deficiency of host signaling protein, Epac1, drives inflammation and glucose intolerance which are the hallmarks of WD-induced obesity. Video abstract.

An artificial intelligence-enabled smartphone app for real-time pressure injury assessment

The management of chronic wounds in the elderly such as pressure injury (also known as bedsore or pressure ulcer) is increasingly important in an ageing population. Accurate classification of the stage of pressure injury is important for wound care planning. Nonetheless, the expertise required for staging is often not available in a residential care home setting. Artificial-intelligence (AI)-based computer vision techniques have opened up opportunities to harness the inbuilt camera in modern smartphones to support pressure injury staging by nursing home carers. In this paper, we summarise the recent development of smartphone or tablet-based applications for wound assessment. Furthermore, we present a new smartphone application (app) to perform real-time detection and staging classification of pressure injury wounds using a deep learning-based object detection system, YOLOv4. Based on our validation set of 144 photos, our app obtained an overall prediction accuracy of 63.2%. The per-class prediction specificity is generally high (85.1%–100%), but have variable sensitivity: 73.3% (stage 1 vs. others), 37% (stage 2 vs. others), 76.7 (stage 3 vs. others), 70% (stage 4 vs. others), and 55.6% (unstageable vs. others). Using another independent test set, 8 out of 10 images were predicted correctly by the YOLOv4 model. When deployed in a real-life setting with two different ambient brightness levels with three different Android phone models, the prediction accuracy of the 10 test images ranges from 80 to 90%, which highlight the importance of evaluation of mobile health (mHealth) application in a simulated real-life setting. This study details the development and evaluation process and demonstrates the feasibility of applying such a real-time staging app in wound care management.

Evaluation of Experimental Protocols for Shotgun Whole-Genome Metagenomic Discovery of Antibiotic Resistance Genes

Shotgun metagenomics has enabled the discovery of antibiotic resistance genes (ARGs). Although there have been numerous studies benchmarking the bioinformatics methods for shotgun metagenomic data analysis, there has not yet been a study that systematically evaluates the performance of different experimental protocols on metagenomic species profiling and ARG detection. In this study, we generated 35 whole genome shotgun metagenomic sequencing data sets for five samples (three human stool and two microbial standard) using seven experimental protocols (KAPA or Flex kits at 50ng, 10ng, or 5ng input amounts; XT kit at 1ng input amount). Using this comprehensive resource, we evaluated the seven protocols in terms of robust detection of ARGs and microbial abundance estimation at various sequencing depths. We found that the data generated by the seven protocols are largely similar. The inter-protocol variability is significantly smaller than the variability between samples or sequencing depths. We found that a sequencing depth of more than 30M is suitable for human stool samples. A higher input amount (50ng) is generally favorable for the KAPA and Flex kits. This systematic benchmarking study sheds light on the impact of sequencing depth, experimental protocol, and DNA input amount on ARG detection in human stool samples.

Aberrant Expression of Circulating MicroRNA Leads to the Dysregulation of Alpha-Synuclein and Other Pathogenic Genes in Parkinson's Disease

A group of circulating microRNAs (miRNAs) have been implicated in the pathogenesis of Parkinson’s disease. However, a comprehensive study of the interactions between pathogenic miRNAs and their downstream Parkinson’s disease (PD)-related target genes has not been performed. Here, we identified the miRNA expression profiles in the plasma and circulating exosomes of Parkinson’s disease patients using next-generation RNA sequencing. Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) analyses showed that the miRNA target genes were enriched in axon guidance, neurotrophin signaling, cellular senescence, and the Transforming growth factor-β (TGF-β), mitogen-activated protein kinase (MAPK), phosphatidylinositol 3-kinase (PI3K)-protein kinase B (AKT) and mechanistic target of rapamycin (mTOR) signaling pathways. Furthermore, a group of aberrantly expressed miRNAs were selected and further validated in individual patient plasma, human neural stem cells (NSCs) and a rat model of PD. More importantly, the full scope of the regulatory network between these miRNAs and their PD-related gene targets in human neural stem cells was examined, and the findings revealed a similar but still varied downstream regulatory cascade involving many known PD-associated genes. Additionally, miR-23b-3p was identified as a novel direct regulator of alpha-synuclein, which is possibly the key component in PD. Our current study, for the first time, provides a glimpse into the regulatory network of pathogenic miRNAs and their PD-related gene targets in PD. Moreover, these PD-associated miRNAs may serve as biomarkers and novel therapeutic targets for PD.

Cisplatin prevents breast cancer metastasis through blocking early EMT and retards cancer growth together with paclitaxel

Cancer growth is usually accompanied by metastasis which kills most cancer patients. Here we aim to study the effect of cisplatin at different doses on breast cancer growth and metastasis.

Methods: We used cisplatin to treat breast cancer cells, then detected the migration of cells and the changes of epithelial-mesenchymal transition (EMT) markers by migration assay, Western blot, and immunofluorescent staining. Next, we analyzed the changes of RNA expression of genes by RNA-seq and confirmed the binding of activating transcription factor 3 (ATF3) to cytoskeleton related genes by ChIP-seq. Thereafter, we combined cisplatin and paclitaxel in a neoadjuvant setting to treat xenograft mouse models. Furthermore, we analyzed the association of disease prognosis with cytoskeletal genes and ATF3 by clinical data analysis.

Results: When administered at a higher dose (6 mg/kg), cisplatin inhibits both cancer growth and metastasis, yet with strong side effects, whereas a lower dose (2 mg/kg) cisplatin blocks cancer metastasis without obvious killing effects. Cisplatin inhibits cancer metastasis through blocking early steps of EMT. It antagonizes transforming growth factor beta (TGFβ) signaling through suppressing transcription of many genes involved in cytoskeleton reorganization and filopodia formation which occur early in EMT and are responsible for cancer metastasis. Mechanistically, TGFβ and fibronectin-1 (FN1) constitute a positive reciprocal regulation loop that is critical for activating TGFβ/SMAD3 signaling, which is repressed by cisplatin induced expression of ATF3. Furthermore, neoadjuvant administration of cisplatin at 2 mg/kg in conjunction with paclitaxel inhibits cancer growth and blocks metastasis without causing obvious side effects by inhibiting colonization of cancer cells in the target organs.

Conclusion: Thus, cisplatin prevents breast cancer metastasis through blocking early EMT, and the combination of cisplatin and paclitaxel represents a promising therapy for killing breast cancer and blocking tumor metastasis.

SeqVis: a tool for detecting compositional heterogeneity among aligned nucleotide sequences

Compositional heterogeneity is a poorly appreciated attribute of aligned nucleotide and amino acid sequences. It is a common property of molecular phylogenetic data, and it has been found to occur across sequences and/or across sites. Most molecular phylogenetic methods assume that the sequences have evolved under globally stationary, reversible, and homogeneous conditions, implying that the sequences should be compositionally homogeneous. The presence of the above-mentioned compositional heterogeneity implies that the sequences must have evolved under more general conditions than is commonly assumed. Consequently, there is a need for reliable methods to detect under what conditions alignments of nucleotides or amino acids may have evolved. In this chapter, we describe one such program. SeqVis is designed to survey aligned nucleotide sequences. We discuss pros-et-cons of this program in the context of other methods to detect compositional heterogeneity and violated phylogenetic assumptions. The benefits provided by SeqVis are demonstrated in two studies of alignments of nucleotides, one of which contained 7542 nucleotides from 53 species.