The Application of Artificial Intelligence in the Analysis of Biomarkers for Diagnosis and Management of Uveitis and Uveal Melanoma: A Systematic Review

Bassi, Arshpreet; Krance, Saffire H.; Pucchio, Aidan; Pur, Daiana R.; Miranda, Rafael N; Felfeli, Tina

doi:10.2147/opth.s377358

Cited by 7 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[49][50][51] Serum biomarkers, including several differentially expressed proteins identi ed in UM gene signatures, have been associated with a worse prognosis in patients diagnosed with UM. 49,52,53 Furthermore, circulating tumor cells have been detected in patients without clinically detectable metastasis, indicating early spread and highlighting the need to identify prognostic biomarkers. 54,55 The search for these biomarkers underscores the importance of early detection of UM, potentially with the aid of ML for the diagnosis and management of UM.…”

Section: Discussionmentioning

confidence: 99%

“…Several studies have evaluated non-imaging biomarkers to better diagnose UM and predict prognosis, including some that employ ML techniques. [49][50][51] Serum biomarkers, including several differentially expressed proteins identi ed in UM gene signatures, have been associated with a worse prognosis in patients diagnosed with UM. 49,52,53 Furthermore, circulating tumor cells have been detected in patients without clinically detectable metastasis, indicating early spread and highlighting the need to identify prognostic biomarkers.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting Malignant Transformation of Choroidal Nevi Using Machine Learning

Iddir,

Love,

et al. 2023

Preprint

View full text Add to dashboard Cite

Objective This study aims to assess a machine learning (ML) algorithm using multimodal imaging to accurately identify risk factors for uveal melanoma (UM) and aid in the diagnosis of melanocytic choroidal tumors. Subjects and Methods This study included 223 eyes from 221 patients with melanocytic choroidal lesions seen at the eye clinic of the University of Illinois at Chicago between 01/2010 and 07/2022. An ML algorithm was developed and trained on ultra-widefield fundus imaging and B-scan ultrasonography to detect risk factors of malignant transformation of choroidal lesions into UM. The risk factors were verified using all multimodal imaging available from the time of diagnosis. We also explore classification of lesions into UM and choroidal nevi using the ML algorithm. Results The ML algorithm assessed features of ultra-widefield fundus imaging and B-scan ultrasonography to determine the presence of the following risk factors for malignant transformation: lesion thickness, subretinal fluid, orange pigment, proximity to optic nerve, ultrasound hollowness, and drusen. The algorithm also provided classification of lesions into UM and choroidal nevi. A total of 115 patients with choroidal nevi and 108 patients with UM were included. The mean lesion thickness for choroidal nevi was 1.6 mm and for UM was 5.9 mm. Eleven ML models were implemented and achieved high accuracy, with an area under the curve of 0.982 for thickness prediction and 0.964 for subretinal fluid prediction. Sensitivity/specificity values ranged from 0.900/0.818 to 1.000/0.727 for different features. The ML algorithm demonstrated high accuracy in identifying risk factors and differentiating lesions based on the analyzed imaging data. Conclusions This study provides proof of concept that ML can accurately identify risk factors for malignant transformation in melanocytic choroidal tumors based on a single ultra-widefield fundus image or B-scan ultrasound at the time of initial presentation. By leveraging the efficiency and availability of ML, this study has the potential to provide a non-invasive tool that helps to prevent unnecessary treatment, improve our ability to predict malignant transformation, reduce the risk of metastasis, and potentially save patient lives.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Predicting Malignant Transformation of Choroidal Nevi Using Machine Learning

Iddir,

Love,

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The use of biomarker analysis with artificial intelligence has been demonstrated in many ophthalmic diseases, such as glaucoma, uveitis, uveal melanoma, age-related macular degeneration, corneal and ocular surface diseases, and retinal occlusive diseases. [8][9][10][11][12] Herein, we evaluated the variability of intraocular inflammatory biomarkers in multiplex assays of intraocular biofluid samples.…”

Section: Introductionmentioning

confidence: 99%

Variability of Replicates of Intraocular Inflammatory Biomarkers in Ocular Fluid Samples Analyzed with Multiplex Assays

Lamoureux,

Wong,

Felfeli

2023

OPTH

Self Cite

View full text Add to dashboard Cite

Certain factors such as instrumental and sample processing errors may contribute to variability of ocular biofluid samples when they are run as replicates with multiplex assays. There is a paucity of literature on the variability of replicates in multiplex assays. This study aims to evaluate whether there is significant variability in replicate analyses of multiplex assays. Methods: A total of 152 human ocular biofluid samples (51 aqueous humor and 101 vitreous) were collected and assayed for 27 cytokine biomarker concentrations (pg/mL). Samples were evaluated as replicates (duplicate analysis) at four different time points. Statistical methods including paired samples t-test, 3-way ANOVA, intraclass correlation coefficient (ICC; <0.5-0.75=poor-moderate, 0.75->0.90 =good-excellent reliability), and coefficients of variation (CV) were employed to evaluate for statistical significance, with Bonferroni corrected P=0.002. Results: Among the 4104 biomarker replicate assays for aqueous humor and vitreous, two analytes (PDGF-BB and IL-7) had a statistically significant difference between the sampled concentrations of the replicates in vitreous samples (mean (diff)=2.05, P<0.001, mean (diff)=1.56, P<0.001, respectively). Majority of the ICC values fell within the good-excellent range (86% of samples) with a minority falling in the poor-moderate range (14% of samples). More variability was noted in the vitreous humour, with five analytes (IL-2, IL-10, IL-12(p70), IL-13, IL-17) demonstrating an average ICC of less than 0.5. The CV calculated for each set of replicates suggested that 93% of replicates had an acceptable level of quantitative assay variability (CV<20%). Conclusion:This study demonstrates that the analysis of most biomarkers in ocular fluids may not require the use of replicates. However, certain analytes such as PDGF-BB and IL-7 may require the use of replicates to ensure reliable results. Caution should be taken when applying these findings to other laboratory settings as our study was conducted by an experienced technician using a standardized protocol. In less standardized settings, replicates may be required in order to ensure accuracy of results. These findings may guide researchers with the design of their studies on ophthalmic biomarker analysis.

show abstract

“…Cytokines are low-molecular-weight glycoprotein molecules produced by both immune and non-immune cells that serve as indicators of normal biological processes, pathogenic processes or pharmacological responses to a therapeutic intervention 1. Recently, there has been a growing interest in using cytokines in aqueous humour and vitreous collected via anterior chamber (AC) paracentesis and vitreous sampling, respectively, as surrogate markers to understand disease pathogenesis, monitor disease progression and assess treatment response 2–12. Given the emerging research and clinical practices based on cytokine analysis, it is imperative to ensure that methodologies to measure cytokines are robust.…”

Section: Introductionmentioning

confidence: 99%

“…1 Recently, there has been a growing interest in using cytokines in aqueous humour and vitreous collected via anterior chamber (AC) paracentesis and vitreous sampling, respectively, as surrogate markers to understand disease pathogenesis, monitor disease progression and assess treatment response. [2][3][4][5][6][7][8][9][10][11][12] Given the emerging research and clinical practices based on cytokine analysis, it is imperative to ensure that methodologies to measure cytokines are robust. Any conclusions drawn from poor handling of the ocular samples and unreliable method of cytokine quantification could potentially result in misled clinical decisions with negative outcomes.…”

Section: Introductionmentioning

confidence: 99%

Evaluating the long-term biological stability of cytokine biomarkers in ocular fluid samples

Felfeli,

Park,

Nestor

et al. 2023

BMJ Open Ophth

Self Cite

View full text Add to dashboard Cite

PurposeThe quality of biological fluid samples is vital for optimal preanalytical procedures and a requirement for effective translational biomarker research. This study aims to determine the effects of storage duration and freeze-thawing on the levels of various cytokines in the human aqueous humour and vitreous samples.Methods and analysisHuman ocular aqueous humour and vitreous samples were obtained from 25 eyes and stored at −80°C for analysis. All samples were assayed for 27 cytokine biomarker concentrations (pg/mL) using a multiplex assay. Four sample storage durations following sample collection were evaluated (1 week, 3 months, 9 months and 15 months). Additionally, samples underwent up to three freeze-thaw cycles within the study period.ResultsAmong the 27 cytokine biomarkers, concentrations of four cytokines (Interleukin (IL)−2, IL-10, IL-12 and platelet-derived growth factor-BB) were significantly decreased by storage duration at all time points, as early as 3 months following sample collection (range of 9%–37% decline between 1 week and 15 months, p<0.001). Freeze-thawing of up to three cycles did not significantly impact the cytokine biomarker concentrations in aqueous humour or vitreous. Separability of patient-specific cytokine biomarker profiles in the principal component analysis remained relatively the same over the 15 months of storage duration.ConclusionThe findings from this study suggest that several intraocular cytokine biomarkers in human aqueous humour and vitreous samples may be susceptible to degradation with long-term storage, as early as 3 months after collection. The overall patient-specific cytokine biomarker profiles are more stable than concentrations of individual cytokines. Future studies should focus on developing guidelines for optimal and standardised sample handling methods to ensure correct research findings about intraocular biomarkers are translated into clinical practice.

show abstract

The Application of Artificial Intelligence in the Analysis of Biomarkers for Diagnosis and Management of Uveitis and Uveal Melanoma: A Systematic Review

Cited by 7 publications

References 28 publications

Predicting Malignant Transformation of Choroidal Nevi Using Machine Learning

Predicting Malignant Transformation of Choroidal Nevi Using Machine Learning

Variability of Replicates of Intraocular Inflammatory Biomarkers in Ocular Fluid Samples Analyzed with Multiplex Assays

Evaluating the long-term biological stability of cytokine biomarkers in ocular fluid samples

Contact Info

Product

Resources

About