Upper gastrointestinal cancers (including oesophageal cancer and gastric cancer) are the most common cancers worldwide. Artificial intelligence platforms using deep learning algorithms have made ...remarkable progress in medical imaging but their application in upper gastrointestinal cancers has been limited. We aimed to develop and validate the Gastrointestinal Artificial Intelligence Diagnostic System (GRAIDS) for the diagnosis of upper gastrointestinal cancers through analysis of imaging data from clinical endoscopies.
This multicentre, case-control, diagnostic study was done in six hospitals of different tiers (ie, municipal, provincial, and national) in China. The images of consecutive participants, aged 18 years or older, who had not had a previous endoscopy were retrieved from all participating hospitals. All patients with upper gastrointestinal cancer lesions (including oesophageal cancer and gastric cancer) that were histologically proven malignancies were eligible for this study. Only images with standard white light were deemed eligible. The images from Sun Yat-sen University Cancer Center were randomly assigned (8:1:1) to the training and intrinsic verification datasets for developing GRAIDS, and the internal validation dataset for evaluating the performance of GRAIDS. Its diagnostic performance was evaluated using an internal and prospective validation set from Sun Yat-sen University Cancer Center (a national hospital) and additional external validation sets from five primary care hospitals. The performance of GRAIDS was also compared with endoscopists with three degrees of expertise: expert, competent, and trainee. The diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of GRAIDS and endoscopists for the identification of cancerous lesions were evaluated by calculating the 95% CIs using the Clopper-Pearson method.
1 036 496 endoscopy images from 84 424 individuals were used to develop and test GRAIDS. The diagnostic accuracy in identifying upper gastrointestinal cancers was 0·955 (95% CI 0·952–0·957) in the internal validation set, 0·927 (0·925–0·929) in the prospective set, and ranged from 0·915 (0·913–0·917) to 0·977 (0·977–0·978) in the five external validation sets. GRAIDS achieved diagnostic sensitivity similar to that of the expert endoscopist (0·942 95% CI 0·924–0·957 vs 0·945 0·927–0·959; p=0·692) and superior sensitivity compared with competent (0·858 0·832–0·880, p<0·0001) and trainee (0·722 0·691–0·752, p<0·0001) endoscopists. The positive predictive value was 0·814 (95% CI 0·788–0·838) for GRAIDS, 0·932 (0·913–0·948) for the expert endoscopist, 0·974 (0·960–0·984) for the competent endoscopist, and 0·824 (0·795–0·850) for the trainee endoscopist. The negative predictive value was 0·978 (95% CI 0·971–0·984) for GRAIDS, 0·980 (0·974–0·985) for the expert endoscopist, 0·951 (0·942–0·959) for the competent endoscopist, and 0·904 (0·893–0·916) for the trainee endoscopist.
GRAIDS achieved high diagnostic accuracy in detecting upper gastrointestinal cancers, with sensitivity similar to that of expert endoscopists and was superior to that of non-expert endoscopists. This system could assist community-based hospitals in improving their effectiveness in upper gastrointestinal cancer diagnoses.
The National Key R&D Program of China, the Natural Science Foundation of Guangdong Province, the Science and Technology Program of Guangdong, the Science and Technology Program of Guangzhou, and the Fundamental Research Funds for the Central Universities.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Background
Due to the occult anatomic location of the nasopharynx and frequent presence of adenoid hyperplasia, the positive rate for malignancy identification during biopsy is low, thus leading to ...delayed or missed diagnosis for nasopharyngeal malignancies upon initial attempt. Here, we aimed to develop an artificial intelligence tool to detect nasopharyngeal malignancies under endoscopic examination based on deep learning.
Methods
An endoscopic images‐based nasopharyngeal malignancy detection model (eNPM‐DM) consisting of a fully convolutional network based on the inception architecture was developed and fine‐tuned using separate training and validation sets for both classification and segmentation. Briefly, a total of 28,966 qualified images were collected. Among these images, 27,536 biopsy‐proven images from 7951 individuals obtained from January 1st, 2008, to December 31st, 2016, were split into the training, validation and test sets at a ratio of 7:1:2 using simple randomization. Additionally, 1430 images obtained from January 1st, 2017, to March 31st, 2017, were used as a prospective test set to compare the performance of the established model against oncologist evaluation. The dice similarity coefficient (DSC) was used to evaluate the efficiency of eNPM‐DM in automatic segmentation of malignant area from the background of nasopharyngeal endoscopic images, by comparing automatic segmentation with manual segmentation performed by the experts.
Results
All images were histopathologically confirmed, and included 5713 (19.7%) normal control, 19,107 (66.0%) nasopharyngeal carcinoma (NPC), 335 (1.2%) NPC and 3811 (13.2%) benign diseases. The eNPM‐DM attained an overall accuracy of 88.7% (95% confidence interval (CI) 87.8%–89.5%) in detecting malignancies in the test set. In the prospective comparison phase, eNPM‐DM outperformed the experts: the overall accuracy was 88.0% (95% CI 86.1%–89.6%) vs. 80.5% (95% CI 77.0%–84.0%). The eNPM‐DM required less time (40 s vs. 110.0 ± 5.8 min) and exhibited encouraging performance in automatic segmentation of nasopharyngeal malignant area from the background, with an average DSC of 0.78 ± 0.24 and 0.75 ± 0.26 in the test and prospective test sets, respectively.
Conclusions
The eNPM‐DM outperformed oncologist evaluation in diagnostic classification of nasopharyngeal mass into benign versus malignant, and realized automatic segmentation of malignant area from the background of nasopharyngeal endoscopic images.
Full text
Available for:
FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SBCE, SBMB, UL, UM, UPUK
It is an important pre-processing step to accurately estimate missing values in microarray data, because complete datasets are required in numerous expression profile analysis in bioinformatics. ...Although several methods have been suggested, their performances are not satisfactory for datasets with high missing percentages.
The paper explores the feasibility of doing missing value imputation with the help of gene regulatory mechanism. An imputation framework called histone acetylation information aided imputation method (HAIimpute method) is presented. It incorporates the histone acetylation information into the conventional KNN(k-nearest neighbor) and LLS(local least square) imputation algorithms for final prediction of the missing values. The experimental results indicated that the use of acetylation information can provide significant improvements in microarray imputation accuracy. The HAIimpute methods consistently improve the widely used methods such as KNN and LLS in terms of normalized root mean squared error (NRMSE). Meanwhile, the genes imputed by HAIimpute methods are more correlated with the original complete genes in terms of Pearson correlation coefficients. Furthermore, the proposed methods also outperform GOimpute, which is one of the existing related methods that use the functional similarity as the external information.
We demonstrated that the using of histone acetylation information could greatly improve the performance of the imputation especially at high missing percentages. This idea can be generalized to various imputation methods to facilitate the performance. Moreover, with more knowledge accumulated on gene regulatory mechanism in addition to histone acetylation, the performance of our approach can be further improved and verified.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Recently, a number of high-resolution genome-wide maps of nucleosome locations in S. cerevisiae have been derived experimentally. However, nucleosome positions are determined in vivo by the combined ...effects of numerous factors. Consequently, nucleosomes are not simple static units, which may explain the discrepancies in reported nucleosome positions as measured by different experiments. In order to more accurately depict the genome-wide nucleosome distribution, we integrated multiple nucleosomal positioning datasets using a multi-angle analysis strategy.
To evaluate the contribution of chromatin structure to transcription, we used the vast amount of available nucleosome analyzed data. Analysis of this data allowed for the comprehensive identification of the connections between promoter nucleosome positioning patterns and various transcription-dependent properties. Further, we characterised the function of nucleosome destabilisation in the context of transcription regulation. Our results indicate that genes with similar nucleosome occupancy patterns share general transcription attributes. We identified the local regulatory correlation (LRC) regions for two distinct types of nucleosomes and we assessed their regulatory properties. We also estimated the nucleosome reproducibility and measurement accuracy for high-confidence transcripts. We found that by maintaining a distance of approximately 13 bp between the upstream border of the +1 nucleosome and the transcription start sites (TSSs), the stable +1 nucleosome may form a barrier against the accessibility of the TSS and shape an optimum chromatin conformation for gene regulation. An in-depth analysis of nucleosome positioning in normally growing and heat shock cells suggested that the extent and patterns of nucleosome sliding are associated with gene activation.
Our results, which combine different types of data, suggest that cross-platform information, including discrepancy and consistency, reflects the mechanisms of nucleosome packaging in vivo more faithfully than individual studies. Furthermore, nucleosomes can be divided into two classes according to their stable and dynamic characteristics. We found that two different nucleosome-positioning characteristics may significantly impact transcription programs. Besides, some positioned-nucleosomes are involved in the transition from stable state to dynamic state in response to abrupt environmental changes.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Neighboring gene pairs in the genome of Saccharomyces cerevisiae have a tendency to be expressed at the same time. The distribution of histone modifications along chromatin fibers is suggested to be ...an important mechanism responsible for such coexpression. However, the extent of the contribution of histone modifications to the coexpression of neighboring genes is unclear.
We investigated the similarity of histone modification between neighboring genes using autocorrelation analysis and composite profiles. Our analysis showed that neighboring genes had similar levels or changes of histone modifications, especially those transcribed in the same direction. The similarities, however, were restricted to 1 or 2 neighboring genes. Moreover, the expression of a gene was significantly correlated with histone modification of its neighboring gene(s), but this was limited to only 1 or 2 neighbors. Using a hidden Markov model (HMM), we found more than 2000 chromatin domains with similar acetylation changes as the cultures changed and a considerable number of these domains covered 2-4 genes. Gene pairs within domains exhibited a higher level of coexpression than random pairs and shared similar functions.
The results of this study suggest that similar histone modifications occur within only a small local chromatin region in yeast. The modifications generally have an effect on coexpression with only 1 or 2 neighboring genes. Some blocking mechanism(s) might strictly restrain the distribution of histone modifications in yeast.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
ATP-dependent chromatin remodeling and the covalent modification of histones play central roles in determining chromatin structure and function. Although several specific interactions between these ...two activities have been elaborated, the global landscape remains to be elucidated.
In this paper, we have developed a computational method to generate the first genome-wide landscape of interactions between ATP-dependent chromatin remodeling and the covalent modification of histones in Saccharomyces cerevisiae. Our method succeeds in identifying known interactions and uncovers many previously unknown interactions between these two activities. Analysis of the genome-wide picture revealed that transcription-related modifications tend to interact with more chromatin remodelers. Our results also demonstrate that most chromatin remodeling-modification interactions act via interactions of remodelers with both histone-modifying enzymes and histone residues. We also found that the co-occurrence of both modification and remodeling has significantly different influences on multiple gene features (e.g. nucleosome occupancy) compared with the presence of either one.
We gave the first genome-wide picture of ATP-dependent chromatin remodeling-histone modification interactions. We also revealed how these two activities work together to regulate chromatin structure and function. Our results suggest that distinct strategies for regulating chromatin activity are selectively employed by genes with different properties.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
•A dual-task AI model was developed to detect and segment NPC in MRI automatically.•The model showed higher overall accuracy than experienced radiologists to detect NPC.•The model showed encouraging ...dice similarity coefficient in segmentation of NPC.
We aimed to develop a dual-task model to detect and segment nasopharyngeal carcinoma (NPC) automatically in magnetic resource images (MRI) based on deep learning method, since the differential diagnosis of NPC and atypical benign hyperplasia was difficult and the radiotherapy target contouring of NPC was labor-intensive.
A self-constrained 3D DenseNet (SC-DenseNet) architecture was improved using separated training and validation sets. A total of 4100 individuals were finally enrolled and split into the training, validation and test sets at a proximate ratio of 8:1:1 using simple randomization. The diagnostic metrics of the established model against experienced radiologists was compared in the test set. The dice similarity coefficient (DSC) of manual and model-defined tumor region was used to evaluate the efficacy of segmentation.
Totally, 3142 nasopharyngeal carcinoma (NPC) and 958 benign hyperplasia were included. The SC-DenseNet model showed encouraging performance in detecting NPC, attained a higher overall accuracy, sensitivity and specificity than those of the experienced radiologists (97.77% vs 95.87%, 99.68% vs 99.24% and 91.67% vs 85.21%, respectively). Moreover, the model also exhibited promising performance in automatic segmentation of tumor region in NPC, with an average DSC at 0.77 ± 0.07 in the test set.
The SC-DenseNet model showed competence in automatic detection and segmentation of NPC in MRI, indicating the promising application value as an assistant tool in clinical practice, especially in screening project.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Nucleosomes regulate DNA accessibility and therefore play a central role in transcription control. Computational methods have been developed to predict static nucleosome positions from DNA sequences, ...but nucleosomes are dynamic in vivo.
Motivated by our observation that transcriptional interaction is discriminative information for nucleosome occupancy, we developed a novel computational approach to identify dynamic nucleosome positions at promoters by combining transcriptional interaction and genomic sequence information. Our approach successfully identified experimentally determined nucleosome positioning dynamics available in three cellular conditions, and significantly improved the prediction accuracy which is based on sequence information alone. We then applied our approach to various cellular conditions and established a comprehensive landscape of dynamic nucleosome positioning in yeast.
Analysis of this landscape revealed that the majority of nucleosome positions are maintained during most conditions. However, nucleosome occupancy at most promoters fluctuates with the corresponding gene expression level and is reduced specifically at the phase of peak expression. Further investigation into properties of nucleosome occupancy identified two gene groups associated with distinct modes of nucleosome modulation. Our results suggest that both the intrinsic sequence and regulatory proteins modulate nucleosomes in an altered manner.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
The nucleosome is the fundamental unit of eukaryotic genomes. Its positioning plays a central role in diverse cellular processes that rely on access to genomic DNA. Experimental evidence suggests ...that the genomic DNA sequence is one important determinant of nucleosome positioning. Yet it is less clear whether the role of the underlying DNA sequence in nucleosome positioning varies across different promoters. Whether different determinants of nucleosome positioning have characteristic influences on nucleosome modulation also remains to be elucidated.
We identified two typical promoter classes in yeast associated with high or low dependence of nucleosome positioning on the underlying DNA sequence, respectively. Importantly, the two classes have low or high intrinsic sequence preferences for nucleosomes, respectively. The two classes are further distinguished by multiple promoter features, including nucleosome occupancy, nucleosome fuzziness, H2A.Z occupancy, changes in nucleosome positions before and after transcriptional perturbation, and gene activity. Both classes have significantly high turnover rates of histone H3, but employ distinct modes of nucleosome modulation: The first class is characterized by hyperacetylation, whereas the second class is highly regulated by ATP-dependent chromatin remodelling.
Our results, coupled with the known features of nucleosome modulation, suggest that the two distinct modes of nucleosome modulation selectively employed by different genes are linked with the intrinsic sequence preferences for nucleosomes. The difference in modes of nucleosome modulation can account for the difference in the contribution of DNA sequence to nucleosome positioning between both promoter classes.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK