Artificial Intelligence in Cataract Surgery: A Systematic Review.

Purpose: The purpose of this study was to assess the current use and reliability of artificial intelligence (AI)-based algorithms for analyzing cataract surgery videos. Methods: A systematic review of the literature about intra-operative analysis of cataract surgery videos with machine learning techniques was performed. Cataract diagnosis and detection algorithms were excluded. Resulting algorithms were compared, descriptively analyzed, and metrics summarized or visually reported. The reproducibility and reliability of the methods and results were assessed using a modified version of the Medical Image Computing and Computer-Assisted (MICCAI) checklist. Results: Thirty-eight of the 550 screened studies were included, 20 addressed the challenge of instrument detection or tracking, 9 focused on phase discrimination, and 8 predicted skill and complications. Instrument detection achieves an area under the receiver operator characteristic curve (ROC AUC) between 0.976 and 0.998, instrument tracking an mAP between 0.685 and 0.929, phase recognition an ROC AUC between 0.773 and 0.990, and complications or surgical skill performs with an ROC AUC between 0.570 and 0.970. Conclusions: The studies showed a wide variation in quality and pose a challenge regarding replication due to a small number of public datasets (none for manual small incision cataract surgery) and seldom published source code. There is no standard for reported outcome metrics and validation of the models on external datasets is rare making comparisons difficult. The data suggests that tracking of instruments and phase detection work well but surgical skill and complication recognition remains a challenge for deep learning. Translational Relevance: This overview of cataract surgery analysis with AI models provides translational value for improving training of the clinician by identifying successes and challenges.

MGCNSS: miRNA-disease association prediction with multi-layer graph convolution and distance-based negative sample selection strategy.

Identifying disease-associated microRNAs (miRNAs) could help understand the deep mechanism of diseases, which promotes the development of new medicine. Recently, network-based approaches have been widely proposed for inferring the potential associations between miRNAs and diseases. However, these approaches ignore the importance of different relations in meta-paths when learning the embeddings of miRNAs and diseases. Besides, they pay little attention to screening out reliable negative samples which is crucial for improving the prediction accuracy. In this study, we propose a novel approach named MGCNSS with the multi-layer graph convolution and high-quality negative sample selection strategy. Specifically, MGCNSS first constructs a comprehensive heterogeneous network by integrating miRNA and disease similarity networks coupled with their known association relationships. Then, we employ the multi-layer graph convolution to automatically capture the meta-path relations with different lengths in the heterogeneous network and learn the discriminative representations of miRNAs and diseases. After that, MGCNSS establishes a highly reliable negative sample set from the unlabeled sample set with the negative distance-based sample selection strategy. Finally, we train MGCNSS under an unsupervised learning manner and predict the potential associations between miRNAs and diseases. The experimental results fully demonstrate that MGCNSS outperforms all baseline methods on both balanced and imbalanced datasets. More importantly, we conduct case studies on colon neoplasms and esophageal neoplasms, further confirming the ability of MGCNSS to detect potential candidate miRNAs. The source code is publicly available on GitHub https://github.com/15136943622/MGCNSS/tree/master.

Tribological Properties of Blocky Composites with Carbon Nanotubes.

A large amount of primary energy is lost due to friction, and the study of new additive materials to improve friction performance is in line with the concept of low carbon. Carbon nanotubes (CNTs) have advantages in drag reduction and wear resistance with their hollow structure and self-lubricating properties. This review investigated the mechanism of improving friction properties of blocky composites (including polymer, metal, and ceramic-based composites) with CNTs' incorporation. The characteristic tubular structure and the carbon film make low wear rate and friction coefficient on the surface. In addition, the effect of CNTs' aggregation and interfacial bond strength on the wear resistance was analyzed. Within an appropriate concentration range of CNTs, the blocky composites exhibit better wear resistance properties. Based on the differences in drag reduction and wear resistance in different materials and preparation methods, further research directions of CNTs have been suggested.

KSNP: a fast de Bruijn graph-based haplotyping tool approaching data-in time cost.

Long reads that cover more variants per read raise opportunities for accurate haplotype construction, whereas the genotype errors of single nucleotide polymorphisms pose great computational challenges for haplotyping tools. Here we introduce KSNP, an efficient haplotype construction tool based on the de Bruijn graph (DBG). KSNP leverages the ability of DBG in handling high-throughput erroneous reads to tackle the challenges. Compared to other notable tools in this field, KSNP achieves at least 5-fold speedup while producing comparable haplotype results. The time required for assembling human haplotypes is reduced to nearly the data-in time.

Application of artificial intelligence-assisted image diagnosis software based on volume data reconstruction technique in medical imaging practice teaching.

BACKGROUND: In medical imaging courses, due to the complexity of anatomical relationships, limited number of practical course hours and instructors, how to improve the teaching quality of practical skills and self-directed learning ability has always been a challenge for higher medical education. Artificial intelligence-assisted diagnostic (AISD) software based on volume data reconstruction (VDR) technique is gradually entering radiology. It converts two-dimensional images into three-dimensional images, and AI can assist in image diagnosis. However, the application of artificial intelligence in medical education is still in its early stages. The purpose of this study is to explore the application value of AISD software based on VDR technique in medical imaging practical teaching, and to provide a basis for improving medical imaging practical teaching. METHODS: Totally 41 students majoring in clinical medicine in 2017 were enrolled as the experiment group. AISD software based on VDR was used in practical teaching of medical imaging to display 3D images and mark lesions with AISD. Then annotations were provided and diagnostic suggestions were given. Also 43 students majoring in clinical medicine from 2016 were chosen as the control group, who were taught with the conventional film and multimedia teaching methods. The exam results and evaluation scales were compared statistically between groups. RESULTS: The total skill scores of the test group were significantly higher compared with the control group (84.51 ± 3.81 vs. 80.67 ± 5.43). The scores of computed tomography (CT) diagnosis (49.93 ± 3.59 vs. 46.60 ± 4.89) and magnetic resonance (MR) diagnosis (17.41 ± 1.00 vs. 16.93 ± 1.14) of the experiment group were both significantly higher. The scores of academic self-efficacy (82.17 ± 4.67) and self-directed learning ability (235.56 ± 13.50) of the group were significantly higher compared with the control group (78.93 ± 6.29, 226.35 ± 13.90). CONCLUSIONS: Applying AISD software based on VDR to medical imaging practice teaching can enable students to timely obtain AI annotated lesion information and 3D images, which may help improve their image reading skills and enhance their academic self-efficacy and self-directed learning abilities.

PySmooth: a Python tool for the removal and correction of genotyping errors.

In genetic mapping studies involving many individuals, genome-wide markers such as single nucleotide polymorphisms (SNPs) can be detected using different methods. However, it comes with some errors. Some SNPs associated with diseases can be in regions encoding long noncoding RNAs (lncRNAs). Therefore, identifying the errors in genotype file and correcting them is crucial for accurate genetic mapping studies. We develop a Python tool called PySmooth, that offers an easy-to-use command line interface for the removal and correction of genotyping errors. PySmooth uses the approach of a previous tool called SMOOTH with some modifications. It inputs a genotype file, detects errors and corrects them. PySmooth provides additional features such as imputing missing data, better user-friendly usage, generates summary and visualization files, has flexible parameters, and handles more genotype codes. AVAILABILITY AND IMPLEMENTATION: PySmooth is available at https://github.com/lncRNAAddict/PySmooth .

A life cycle analysis of the environmental impact of procurement, waste and water in the dental practice.

Background Health care is a significant contributor to climate change. Global pressure for a change towards a more sustainable way of providing dental health care has resulted in the creation of the Green Impact Toolkit, which is comprised of a list of suggested changes that dental practices can make to become more sustainable in a number of categories, such as procurement, waste and water.Aims To compare the effectiveness of changes suggested by the Green Impact Toolkit.Materials and methods A comparative life cycle assessment (LCA) was conducted using the Ecoinvent database v3.8 and these data were processed using OpenLCA v1.10.3 software.Results The carbon footprint per patient was significantly reduced after the recommendations were implemented. For instance, using water from a rainwater collection tank instead of the mains supply saved 30 g CO2eq (carbon dioxide equivalents) per patient, a 90% reduction in carbon footprint.Discussion This comparative LCA identified some effective changes which can be easily made by a dental practice. Nevertheless, some actions require some initial financial investment and may be difficult to implement in a busy modern dental practice setting.Conclusion The findings from this study can be used to guide dental practices to making choices which are more sustainable and eco-friendly in the future.

Study of association between corneal shape parameters and axial length elongation during orthokeratology using image-pro plus software.

BACKGROUND: The aim was to validate the correlation between corneal shape parameters and axial length growth (ALG) during orthokeratology using Image-Pro Plus (IPP) 6.0 software. METHODS: This retrospective study used medical records of myopic children aged 8-13 years (n = 104) undergoing orthokeratology. Their corneal topography and axial length were measured at baseline and subsequent follow-ups after lens wear. Corneal shape parameters, including the treatment zone (TZ) area, TZ diameter, TZ fractal dimension, TZ radius ratio, eccentric distance, pupil area, and pupillary peripheral steepened zone(PSZ) area, were measured using IPP software. The impact of corneal shape parameters at 3 months post-orthokeratology visit on 1.5-year ALG was evaluated using multivariate linear regression analysis. RESULTS: ALG exhibited significant associations with age, TZ area, TZ diameter, TZ fractal dimension, and eccentric distance on univariate linear regression analysis. Multivariate regression analysis identified age, TZ area, and eccentric distance as significantly correlated with ALG (all P < 0.01), with eccentric distance showing the strongest correlation (ß = -0.370). The regressive equation was y = 1.870 - 0.235a + 0.276b - 0.370c, where y represents ALG, a represents age, b represents TZ area, and c represents eccentric distance; R2 = 0.27). No significant relationships were observed between the TZ radius ratio, pupillary PSZ area, and ALG. CONCLUSIONS: IPP software proves effective in capturing precise corneal shape parameters after orthokeratology. Eccentric distance, rather than age or the TZ area, significantly influences ALG retardation.

Synthetic Iris Images: A Comparative Analysis between Cartesian and Polar Representation.

In recent years, the advancement of generative techniques, particularly generative adversarial networks (GANs), has opened new possibilities for generating synthetic biometric data from different modalities, including-among others-images of irises, fingerprints, or faces in different representations. This study presents the process of generating synthetic images of human irises, using the recent StyleGAN3 model. The novelty presented in this work consists in producing generated content in both Cartesian and polar coordinate representations, typically used in iris recognition pipelines, such as the foundational work proposed by John Daugman, but hitherto not used in generative AI experiments. The main objective of this study was to conduct a qualitative analysis of the synthetic samples and evaluate the iris texture density and suitability for meaningful feature extraction. During this study, a total of 1327 unique irises were generated, and experimental results carried out using the well-known OSIRIS open-source iris recognition software and the equivalent software, wordlcoin-openiris, newly published at the end of 2023 to prove that (1) no "identity leak" from the training set was observed, and (2) the generated irises had enough unique textural information to be successfully differentiated between both themselves and between them and real, authentic iris samples. The results of our research demonstrate the promising potential of synthetic iris data generation as a valuable tool for augmenting training datasets and improving the overall performance of iris recognition systems. By exploring the synthetic data in both Cartesian and polar representations, we aim to understand the benefits and limitations of each approach and their implications for biometric applications. The findings suggest that synthetic iris data can significantly contribute to the advancement of iris recognition technology, enhancing its accuracy and robustness in real-world scenarios by greatly augmenting the possibilities to gather large and diversified training datasets.

MSK-TIM: A Telerobotic Ultrasound System for Assessing the Musculoskeletal System.

The aim of this paper is to investigate technological advancements made to a robotic tele-ultrasound system for musculoskeletal imaging, the MSK-TIM (Musculoskeletal Telerobotic Imaging Machine). The hardware was enhanced with a force feedback sensor and a new controller was introduced. Software improvements were developed which allowed the operator to access ultrasound functions such as focus, depth, gain, zoom, color, and power Doppler controls. The device was equipped with Wi-Fi network capability which allowed the master and slave stations to be positioned in different locations. A trial assessing the system to scan the wrist was conducted with twelve participants, for a total of twenty-four arms. Both the participants and radiologist reported their experience. The images obtained were determined to be of satisfactory quality for diagnosis. The system improvements resulted in a better user and patient experience for the radiologist and participants. Latency with the VPN configuration was similar to the WLAN in our experiments. This research explores several technologies in medical telerobotics and provides insight into how they should be used in future. This study provides evidence to support larger-scale trials of the MSK-TIM for musculoskeletal imaging.

Using Computer Vision to Annotate Video-Recoded Direct Observation of Physical Behavior.

Direct observation is a ground-truth measure for physical behavior, but the high cost limits widespread use. The purpose of this study was to develop and test machine learning methods to recognize aspects of physical behavior and location from videos of human movement: Adults (N = 26, aged 18-59 y) were recorded in their natural environment for two, 2- to 3-h sessions. Trained research assistants annotated videos using commercially available software including the following taxonomies: (1) sedentary versus non-sedentary (two classes); (2) activity type (four classes: sedentary, walking, running, and mixed movement); and (3) activity intensity (four classes: sedentary, light, moderate, and vigorous). Four machine learning approaches were trained and evaluated for each taxonomy. Models were trained on 80% of the videos, validated on 10%, and final accuracy is reported on the remaining 10% of the videos not used in training. Overall accuracy was as follows: 87.4% for Taxonomy 1, 63.1% for Taxonomy 2, and 68.6% for Taxonomy 3. This study shows it is possible to use computer vision to annotate aspects of physical behavior, speeding up the time and reducing labor required for direct observation. Future research should test these machine learning models on larger, independent datasets and take advantage of analysis of video fragments, rather than individual still images.

POCP-nf: an automatic Nextflow pipeline for calculating the percentage of conserved proteins in bacterial taxonomy.

SUMMARY: Sequence technology advancements have led to an exponential increase in bacterial genomes, necessitating robust taxonomic classification methods. The Percentage Of Conserved Proteins (POCP), proposed initially by Qin et al. (2014), is a valuable metric for assessing prokaryote genus boundaries. Here, I introduce a computational pipeline for automated POCP calculation, aiming to enhance reproducibility and ease of use in taxonomic studies. AVAILABILITY AND IMPLEMENTATION: The POCP-nf pipeline uses DIAMOND for faster protein alignments, achieving similar sensitivity to BLASTP. The pipeline is implemented in Nextflow with Conda and Docker support and is freely available on GitHub under https://github.com/hoelzer/pocp. The open-source code can be easily adapted for various prokaryotic genome and protein datasets. Detailed documentation and usage instructions are provided in the repository.

Slimmable transformer with hybrid axial-attention for medical image segmentation.

The transformer architecture has achieved remarkable success in medical image analysis owing to its powerful capability for capturing long-range dependencies. However, due to the lack of intrinsic inductive bias in modeling visual structural information, the transformer generally requires a large-scale pre-training schedule, limiting the clinical applications over expensive small-scale medical data. To this end, we propose a slimmable transformer to explore intrinsic inductive bias via position information for medical image segmentation. Specifically, we empirically investigate how different position encoding strategies affect the prediction quality of the region of interest (ROI) and observe that ROIs are sensitive to different position encoding strategies. Motivated by this, we present a novel Hybrid Axial-Attention (HAA) that can be equipped with pixel-level spatial structure and relative position information as inductive bias. Moreover, we introduce a gating mechanism to achieve efficient feature selection and further improve the representation quality over small-scale datasets. Experiments on LGG and COVID-19 datasets prove the superiority of our method over the baseline and previous works. Internal workflow visualization with interpretability is conducted to validate our success better; the proposed slimmable transformer has the potential to be further developed into a visual software tool for improving computer-aided lesion diagnosis and treatment planning.

Salivary and serum levels of lactate dehydrogenase in oral submucous fibrosis: A meta-analysis.

BACKGROUND: The occurrence of oral submucous fibrosis (OSF) is often accompanied by an increase in lactate dehydrogenase (LDH) levels. In this meta-analysis, we compared the salivary and serum levels of LDH levels between OSF patients and controls. MATERIAL AND METHODS: A comprehensive search was conducted in PubMed, Embase, Web of Science, and Cochrane Library from the establishment of the database to June 2023, and the quality of the studies was checked by the Newcastle-Ottawa Quality Assessment scale. The mean difference (MD) and 95% confidence interval (CI) were calculated using RevMan 5.4 software. RESULTS: A total of 28 studies were retrieved from the database, and we included 5 studies in this meta-analysis. The salivary LDH level of OSF patients was higher than healthy controls (MD: 423.10 pg/L 95%CI: 276.42-569.77 pg/mL, P < .00001), the serum LDH level of OSF patients was also higher than that of healthy controls (MD: 226.20 pg/mL, 95%CI: 147.71-304.69 pg/mL, P < .00001). CONCLUSIONS: This meta-analysis showed that salivary and serum LDH levels were higher in OSF patients than in healthy controls, suggesting that LDH may be a potential biomarker for OSF.

Automated 2D and 3D finite element overclosure adjustment and mesh morphing using generalized regression neural networks.

Computer representations of three-dimensional (3D) geometries are crucial for simulating systems and processes in engineering and science. In medicine, and more specifically, biomechanics and orthopaedics, obtaining and using 3D geometries is critical to many workflows. However, while many tools exist to obtain 3D geometries of organic structures, little has been done to make them usable for their intended medical purposes. Furthermore, many of the proposed tools are proprietary, limiting their use. This work introduces two novel algorithms based on Generalized Regression Neural Networks (GRNN) and 4 processes to perform mesh morphing and overclosure adjustment. These algorithms were implemented, and test cases were used to validate them against existing algorithms to demonstrate improved performance. The resulting algorithms demonstrate improvements to existing techniques based on Radial Basis Function (RBF) networks by converting to GRNN-based implementations. Implementations in MATLAB of these algorithms and the source code are publicly available at the following locations: https://github.com/thor-andreassen/femors; https://simtk.org/projects/femors-rbf; https://www.mathworks.com/matlabcentral/fileexchange/120353-finite-element-morphing-overclosure-reduction-and-slicing.

3Dmapper: a command line tool for BioBank-scale mapping of variants to protein structures.

MOTIVATION: The interpretation of genomic data is crucial to understand the molecular mechanisms of biological processes. Protein structures play a vital role in facilitating this interpretation by providing functional context to genetic coding variants. However, mapping genes to proteins is a tedious and error-prone task due to inconsistencies in data formats. Over the past two decades, numerous tools and databases have been developed to automatically map annotated positions and variants to protein structures. However, most of these tools are web-based and not well-suited for large-scale genomic data analysis. RESULTS: To address this issue, we introduce 3Dmapper, a stand-alone command-line tool developed in Python and R. It systematically maps annotated protein positions and variants to protein structures, providing a solution that is both efficient and reliable. AVAILABILITY AND IMPLEMENTATION: https://github.com/vicruiser/3Dmapper.

A machine-readable specification for genomics assays.

MOTIVATION: Understanding the structure of sequenced fragments from genomics libraries is essential for accurate read preprocessing. Currently, different assays and sequencing technologies require custom scripts and programs that do not leverage the common structure of sequence elements present in genomics libraries. RESULTS: We present seqspec, a machine-readable specification for libraries produced by genomics assays that facilitates standardization of preprocessing and enables tracking and comparison of genomics assays. AVAILABILITY AND IMPLEMENTATION: The specification and associated seqspec command line tool is available at https://www.doi.org/10.5281/zenodo.10213865.

Advanced Tandem Mass Spectrometric Analysis of Complex Mixtures of Triacylglycerol Regioisomers: A Case Study of Bovine Milk Fat.

Comprehensive analysis of triacylglycerol (TAG) regioisomers is extremely challenging, with many variables that can influence the results. Previously, we reported a novel algorithmic method for resolving regioisomers of complex mixtures of TAGs. In the current study, the TAG Analyzer software and its mass spectrometric fragmentation model were further developed and validated for a much wider range of TAGs. To demonstrate the method, we performed for the first time a comprehensive analysis of TAG regioisomers of bovine milk fat, a very important and one of the most complex TAG mixtures in nature containing FAs ranging from short to long carbon chains. This analysis method forms a solid basis for further investigation of TAG regioisomer profiles in various natural fats and oils, potentially aiding in the development of new and healthier foods and nutraceuticals with targeted lipid structures.

CITEViz: interactively classify cell populations in CITE-Seq via a flow cytometry-like gating workflow using R-Shiny.

BACKGROUND: The rapid advancement of new genomic sequencing technology has enabled the development of multi-omic single-cell sequencing assays. These assays profile multiple modalities in the same cell and can often yield new insights not revealed with a single modality. For example, Cellular Indexing of Transcriptomes and Epitopes by Sequencing (CITE-Seq) simultaneously profiles the RNA transcriptome and the surface protein expression. The surface protein markers in CITE-Seq can be used to identify cell populations similar to the iterative filtration process in flow cytometry, also called "gating", and is an essential step for downstream analyses and data interpretation. While several packages allow users to interactively gate cells, they often do not process multi-omic sequencing datasets and may require writing redundant code to specify gate boundaries. To streamline the gating process, we developed CITEViz which allows users to interactively gate cells in Seurat-processed CITE-Seq data. CITEViz can also visualize basic quality control (QC) metrics allowing for a rapid and holistic evaluation of CITE-Seq data. RESULTS: We applied CITEViz to a peripheral blood mononuclear cell CITE-Seq dataset and gated for several major blood cell populations (CD14 monocytes, CD4 T cells, CD8 T cells, NK cells, B cells, and platelets) using canonical surface protein markers. The visualization features of CITEViz were used to investigate cellular heterogeneity in CD14 and CD16-expressing monocytes and to detect differential numbers of detected antibodies per patient donor. These results highlight the utility of CITEViz to enable the robust classification of single cell populations. CONCLUSIONS: CITEViz is an R-Shiny app that standardizes the gating workflow in CITE-Seq data for efficient classification of cell populations. Its secondary function is to generate basic feature plots and QC figures specific to multi-omic data. The user interface and internal workflow of CITEViz uniquely work together to produce an organized workflow and sensible data structures for easy data retrieval. This package leverages the strengths of biologists and computational scientists to assess and analyze multi-omic single-cell datasets. In conclusion, CITEViz streamlines the flow cytometry gating workflow in CITE-Seq data to help facilitate novel hypothesis generation.