Comparison of Chest Radiograph Interpretations by Artificial Intelligence Algorithm vs Radiology Residents.

Importance: Chest radiography is the most common diagnostic imaging examination performed in emergency departments (EDs). Augmenting clinicians with automated preliminary read assistants could help expedite their workflows, improve accuracy, and reduce the cost of care. Objective: To assess the performance of artificial intelligence (AI) algorithms in realistic radiology workflows by performing an objective comparative evaluation of the preliminary reads of anteroposterior (AP) frontal chest radiographs performed by an AI algorithm and radiology residents. Design, Setting, and Participants: This diagnostic study included a set of 72 findings assembled by clinical experts to constitute a full-fledged preliminary read of AP frontal chest radiographs. A novel deep learning architecture was designed for an AI algorithm to estimate the findings per image. The AI algorithm was trained using a multihospital training data set of 342 126 frontal chest radiographs captured in ED and urgent care settings. The training data were labeled from their associated reports. Image-based F1 score was chosen to optimize the operating point on the receiver operating characteristics (ROC) curve so as to minimize the number of missed findings and overcalls per image read. The performance of the model was compared with that of 5 radiology residents recruited from multiple institutions in the US in an objective study in which a separate data set of 1998 AP frontal chest radiographs was drawn from a hospital source representative of realistic preliminary reads in inpatient and ED settings. A triple consensus with adjudication process was used to derive the ground truth labels for the study data set. The performance of AI algorithm and radiology residents was assessed by comparing their reads with ground truth findings. All studies were conducted through a web-based clinical study application system. The triple consensus data set was collected between February and October 2018. The comparison study was preformed between January and October 2019. Data were analyzed from October to February 2020. After the first round of reviews, further analysis of the data was performed from March to July 2020. Main Outcomes and Measures: The learning performance of the AI algorithm was judged using the conventional ROC curve and the area under the curve (AUC) during training and field testing on the study data set. For the AI algorithm and radiology residents, the individual finding label performance was measured using the conventional measures of label-based sensitivity, specificity, and positive predictive value (PPV). In addition, the agreement with the ground truth on the assignment of findings to images was measured using the pooled κ statistic. The preliminary read performance was recorded for AI algorithm and radiology residents using new measures of mean image-based sensitivity, specificity, and PPV designed for recording the fraction of misses and overcalls on a per image basis. The 1-sided analysis of variance test was used to compare the means of each group (AI algorithm vs radiology residents) using the F distribution, and the null hypothesis was that the groups would have similar means. Results: The trained AI algorithm achieved a mean AUC across labels of 0.807 (weighted mean AUC, 0.841) after training. On the study data set, which had a different prevalence distribution, the mean AUC achieved was 0.772 (weighted mean AUC, 0.865). The interrater agreement with ground truth finding labels for AI algorithm predictions had pooled κ value of 0.544, and the pooled κ for radiology residents was 0.585. For the preliminary read performance, the analysis of variance test was used to compare the distributions of AI algorithm and radiology residents' mean image-based sensitivity, PPV, and specificity. The mean image-based sensitivity for AI algorithm was 0.716 (95% CI, 0.704-0.729) and for radiology residents was 0.720 (95% CI, 0.709-0.732) (P = .66), while the PPV was 0.730 (95% CI, 0.718-0.742) for the AI algorithm and 0.682 (95% CI, 0.670-0.694) for the radiology residents (P < .001), and specificity was 0.980 (95% CI, 0.980-0.981) for the AI algorithm and 0.973 (95% CI, 0.971-0.974) for the radiology residents (P < .001). Conclusions and Relevance: These findings suggest that it is possible to build AI algorithms that reach and exceed the mean level of performance of third-year radiology residents for full-fledged preliminary read of AP frontal chest radiographs. This diagnostic study also found that while the more complex findings would still benefit from expert overreads, the performance of AI algorithms was associated with the amount of data available for training rather than the level of difficulty of interpretation of the finding. Integrating such AI systems in radiology workflows for preliminary interpretations has the potential to expedite existing radiology workflows and address resource scarcity while improving overall accuracy and reducing the cost of care.

Modification of histone by glyoxal: recognition of glycated histone containing advanced glycation adducts by serum antibodies of type 1 diabetes patients.

Dicarbonyl compounds react more rapidly, than glucose, with arginine and lysine in proteins to form advanced glycation end products (AGEs) and further produce free radicals which cause DNA damage. AGEs are reliable diagnostic biomarkers for most of the age-related diseases. In the present study histone was modified with glyoxal and it was characterized by various spectral techniques. Binding characteristics of the modified histone towards serum antibodies from type 1 diabetes patients was evaluated by solid phase enzyme immunoassay and the results were compared with normal human subjects. Fluorescence and Fourier transformed infrared analysis of the nuclear protein clearly indicated changes in their respective intensities upon modification with glyoxal. Liquid chromatography together with mass spectrometry showed new peaks and m/z values related to AGE adducts of dihydroimidazolidines/hydroimidazolones. This glyoxal modified protein was recognized by serum antibodies of the diabetes patients while it showed negligible binding with that of normal human subjects. Glyoxal modification of histone causes structural turbulence and formation of advanced glycation adducts in histone. These adducts might be the main antigenic epitope of the modified histone, leading to its recognition by circulating type 1 diabetes antibodies.

Immuno-chemistry of hydroxyl radical modified GAD-65: A possible role in experimental and human diabetes mellitus.

The repertoire of known auto-antigens is limited to a very small proportion of all human proteins, and the reason why only some proteins become auto-antigens is unclear. The 65 kDa isoform of the enzyme glutamic acid decarboxylase (GAD-65) is a major auto-antigen in type I diabetes, and in various neurological diseases. Most patients with type I diabetes (70-80%) have auto-antibodies against GAD-65, which often appear years before clinical onset of the autoimmune diabetes. Thus, the aim of the study is to focus on the immunogenicity of GAD65 and its reactive oxygen species (ROS) conformer in STZ-induced diabetic rats and on human diabetic patients. In the present study, GAD-65 was modified by hydroxyl radical following Fenton's reaction. The modifications in the structure of the GAD-65 are supported by UV-vis and fluorescence spectral studies. Immunogenicity of both native and hydroxyl radical modified GAD-65 (ROS-GAD-65) was studied in experimental rabbits and was confirmed by inducing type I diabetes in experimental male albino rats using streptozotocin (45 mg/kg). We found that ROS-GAD-65 was a better immunogen as compared to the native GAD-65. A considerable high binding to ROS-GAD-65 was observed as compared to native GAD-65 in both the serum antibodies from diabetes animal models and as well as in the serum samples of type I diabetes. Hydrogen peroxide under the exposure of UV light produces hydroxyl radical (·OH) which is most potent oxidant, and could cause protein damage (GAD-65) to the extent of generating neo-epitopes on the molecule, thus making it immunogenic.

PR3ANCA Related Cerebral Vasculitis in Ulcerative Colitis Presenting with Orbital Involvement: A Case Report with Review of Literature.

PR3 ANCA is a classic marker of granulomatosis with polyangiitis (GPA). There have been several recent reports of increased prevalence of PR3ANCA in ulcerative colitis (UC) patients, the clinical implication of which is not well defined. We are reporting a case of 27-year-old Caucasian male with 14-year history of UC presenting with unilateral proptosis, conjunctival congestion, and chemosis who developed acute hemiparesis within three days of hospital admission, followed by rapid neurological deterioration correlating with brain imaging findings. Serologically he had atypical PANCA with high PR3 antibody titer with a negative infectious workup. His cerebral angiogram was normal but the brain biopsy showed necrotizing vasculitis. He was diagnosed with PR3 ANCA mediated cerebral and orbital vasculitis associated with UC. Treatment was initiated with high dose steroids, plasmapheresis, and cyclophosphamide. He improved significantly with residual left hemiparesis.

Role of early glycation Amadori products of lysine-rich proteins in the production of autoantibodies in diabetes type 2 patients.

In diabetes, protein glycation mostly occurs at intrachain lysine residues resulting in the formation of early stage Amadori products which are finally converted to advance glycation end products (AGEs). Several studies have reported autoantibodies against AGEs in diabetes but not much data are found in respect of Amadori products. In this study, poly-L-lysine (PLL) was glycated with 50 mM glucose and the resultant Amadori products were estimated by fructosamine or nitroblue tetrazolium assay. We report high content of Amadori products in PLL upon glycation. Glycated PLL showed marked hyperchromicity in the UV spectrum, ellipticity changes in CD spectroscopy, and variations in ε-methylene protons shift in NMR. It was better recognized by autoantibodies in type 2 diabetics compared to the native PLL. Induced antibodies against glycated PLL were successfully used to probe early glycation in the IgG isolated from diabetes type 2 patients. Role of Amadori products of glycated proteins in the induction of autoantibodies in type 2 diabetes as well as in associated secondary complications has been discussed.

Amadori glycated proteins: role in production of autoantibodies in diabetes mellitus and effect of inhibitors on non-enzymatic glycation.

Nonenzymatic glycation of macromolecules, especially proteins leading to their oxidation is increased in diabetes mellitus due to hyperglycemia and play an important role in associated complications of the disease. The glycation primarily occurs at intrachain lysine residues of proteins and results in the formation of an early stage stable product as Amadori-lysine which undergo further irreversible chemical reactions to form advanced glycation endproducts. This review deals with the role of Amadori modified proteins in pathogenesis of diabetes. We aim to explain immunogenicity of Amadori-glycated proteins, which might be involve in production of serum autoantibodies in the diabetic patients, and effect of inhibitors on the glycation process.

Biochemical Studies on Methylglyoxal-Mediated Glycated Histones: Implications for Presence of Serum Antibodies against the Glycated Histones in Patients with Type 1 Diabetes Mellitus.

Reactive carbonyl species (RCS) mainly reacts with lysine and arginine residues of proteins to form advanced glycation end products (AGEs). Histone was glycoxidated with glyoxal and methylglyoxal. It was characterized by polyacrylamide gel electrophoresis and quenching studies involving penicillamine and aminoguanidine as carbonyl scavengers. Further characterization of histone modified with methylglyoxal was done by UV, fluorescence, and IR spectrophotometry. Spectral analysis of the protein clearly demonstrates structural perturbation in the histone by methylglyoxal. Methylglyoxal-induces cross-linking in the protein leading to aggregation. Role of methylglyoxal mediated glycoxidation of histone in type 1 diabetes was also undertaken. Antibodies were detected against glycoxidated histone in sera of type 1 diabetes patients by solid-phase enzyme immunoassay. The findings indicate that as a result of structural perturbation in histone by methylglyoxal, the modified histone may be involved in production of serum antibodies in the diabetes patients.

Detecting phenotype-specific interactions between biological processes from microarray data and annotations.

High throughput technologies enable researchers to measure expression levels on a genomic scale. However, the correct and efficient biological interpretation of such voluminous data remains a challenging problem. Many tools have been developed for the analysis of GO terms that are over- or under-represented in a list of differentially expressed genes. However, a previously unexplored aspect is the identification of changes in the way various biological processes interact in a given condition with respect to a reference. Here, we present a novel approach that aims at identifying such interactions between biological processes that are significantly different in a given phenotype with respect to normal. The proposed technique uses vector-space representation, SVD-based dimensionality reduction, differential weighting, and bootstrapping to asses the significance of the interactions under the multiple and complex dependencies expected between the biological processes. We illustrate our approach on two real data sets involving breast and lung cancer. More than 88 percent of the interactions found by our approach were deemed to be correct by an extensive manual review of literature. An interesting subset of such interactions is discussed in detail and shown to have the potential to open new avenues for research in lung and breast cancer.

Glycated lysine residues: a marker for non-enzymatic protein glycation in age-related diseases.

Nonenzymatic glycosylation or glycation of macromolecules, especially proteins leading to their oxidation, play an important role in diseases. Glycation of proteins primarily results in the formation of an early stage and stable Amadori-lysine product which undergo further irreversible chemical reactions to form advanced glycation endproducts (AGEs). This review focuses these products in lysine rich proteins such as collagen and human serum albumin for their role in aging and age-related diseases. Antigenic characteristics of glycated lysine residues in proteins together with the presence of serum autoantibodies to the glycated lysine products and lysine-rich proteins in diabetes and arthritis patients indicates that these modified lysine residues may be a novel biomarker for protein glycation in aging and age-related diseases.

Physicochemical analysis of poly-L-lysine: An insight into the changes induced in lysine residues of proteins on modification with glucose.

Nonenzymatic glycation of macromolecules, especially proteins, takes place mainly due to hyperglycemia in diabetes mellitus. Increased glycolysis during cancer and inflammation during rheumatoid arthritis also contribute to the process of glycation. As lysine residues of proteins are a potential site for glycation, it could be used as a marker for early glycation induced changes in lysine-rich proteins. In the present study, a lysine polymer was incubated with increasing concentrations of glucose for 24 h, and the early glycation product was evaluated by nitroblue tetrazolium assay. The modified polymer together with unmodified one was characterized by gel electrophoresis and UV, fluorescence spectroscopy. Results of the study clearly demonstrate that structural perturbation in the lysine polymer was caused by the early glycation. Further study on detection of antibodies against theglycated proteins in diseased patients might be helpful in early diagnosis of the disease.

Preferential recognition of Amadori-rich lysine residues by serum antibodies in diabetes mellitus: role of protein glycation in the disease process.

This study analyzes effect of glycation on proteins rich in lysine residues as hyperglycemia induced protein glycation has been mainly reported in diabetes mellitus at the intrachain lysine residues leading to the formation of Amadori modified proteins. We have studied the effect of glucose on poly-l-lysine (PLL), a homopolymer of lysine residues. Levels of Amadori products in the glycated PLL were evaluated by fructosamine assay and the presence of 5-hydroxymethylfurfural (HMF) in the glycated PLL was analyzed by thiobarbituric acid assay. Fluorescence and FT-IR spectroscopy were applied to characterize the modified PLL. Binding characteristics of experimentally induced antibodies against glycated PLL and the presence of antibodies against glycated PLL in the sera of diabetes patients was evaluated by solid phase enzyme immunoassays. The fructosamine assay showed significantly high yield of early glycation (Amadori) products in the glycated PLL, which was confirmed by increased yield of HMF from Amadori products of glycated PLL. Loss in fluorescence intensity and appearance of a new band corresponding to Amadori products were observed in FT-IR spectrum of the glycated PLL. Glycated PLL was found to be highly immunogenic in rabbits as compared to the native form. Serum antibodies from diabetes patients showed appreciably high recognition of the glycated PLL. The results conclusively show the glycation induced damage to the lysine molecules and specific recognition of Amadori-lysine residues by serum antibodies from diabetes patients. The glycated lysine residues may serve as a diagnostic biomarker for early glycation process in diabetes mellitus.

Onto-Tools: new additions and improvements in 2006.

Onto-Tools is a freely available web-accessible software suite, composed of an annotation database and nine complementary data-mining tools. This article describes a new tool, Onto-Express-to-go (OE2GO), as well as some new features implemented in Pathway-Express and Onto-Miner over the past year. Pathway-Express (PE) has been enhanced to identify significantly perturbed pathways in a given condition using the differentially expressed genes in the input. OE2GO is a tool for functional profiling using custom annotations. The development of this tool was aimed at the researchers working with organisms for which annotations are not yet available in the public domain. OE2GO allows researchers to use either annotation data from the Onto-Tools database, or their own custom annotations. By removing the necessity to use any specific database, OE2GO makes the functional profiling available for all organisms, with annotations using any ontology. The Onto-Tools are freely available at http://vortex.cs.wayne.edu/projects.htm.

Immunoglobulin heavy and light chain isotypes in multiple myeloma patients.

The frequency of expression of immunoglobulin (Ig) light and heavy chain isotypes was analyzed in myeloma proteins (M-proteins) from sera of 40 Indian patients with clinically established multiple myeloma. Patients samples were screened by a combination of electrophoresis, immunoelectrophoresis (IEP) and ELISA techniques in this study. We found that majority of the myeloma proteins (58%) were of the IgG isotype followed by IgA (24%) and biclonal gammopathy associated with IgG and IgA (5%). Both kappa and lambda light chains were associated with the heavy chain isotypes. We recommend the triangular combination for detection of M-proteins and biclonal gammopathy of cancerous plasma cells as biomarkers for diagnosis of myeloma.