Unsupervised Machine Learning Identifies Quantifiable Patterns of Visual Field Loss in Idiopathic Intracranial Hypertension.
Ontology highlight
ABSTRACT: Purpose
Archetypal analysis, a form of unsupervised machine learning, identifies archetypal patterns within a visual field (VF) dataset such that any VF is described as a weighted sum of its archetypes (ATs) and has been used to quantify VF defects in glaucoma. We applied archetypal analysis to VFs affected by nonglaucomatous optic neuropathy caused by idiopathic intracranial hypertension (IIH).Methods
We created an AT model from 2862 VFs prospectively collected from 330 eyes in the IIH Treatment Trial (IIHTT). We compared baseline IIH AT patterns with their descriptive VF classifications from the IIHTT.Results
The optimum IIH AT model yielded 14 ATs resembling VF patterns reported in the IIHTT. Baseline VFs contained four or fewer meaningful ATs in 147 (89%) of study eyes. AT2 (mild general VF depression pattern) demonstrated the greatest number of study eyes with meaningful AT weight at baseline (n = 114), followed by AT1 (n = 91). Other ATs captured patterns of blind spot enlargement, hemianopia, arcuate, nasal defects, and more nonspecific patterns of general VF depression. Of all ATs, AT1 (normal pattern) had the strongest correlation with mean deviation (r = 0.69, P < 0.001). For 65 of the 93 VFs with a dominant AT, this AT matched the expert classification.Conclusions
Archetypal analysis identifies quantifiable, archetypal VF defects that resemble those commonly seen in IIH.Translational relevance
Archetypal analysis provides a quantitative, objective method of measuring and monitoring disease-specific regional VF defects in IIH.
SUBMITTER: Doshi H
PROVIDER: S-EPMC8411857 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:ImportanceAcetazolamide is commonly used to treat idiopathic intracranial hypertension (IIH), but there is insufficient information to establish an evidence base for its use.ObjectiveTo determine whether acetazolamide is beneficial in improving vision when added to a low-sodium weight reduction diet in patients with IIH and mild visual loss.Design, setting, and participantsMulticenter, randomized, double-masked, placebo-controlled study of acetazolamide in 165 participants with IIH and mild visual loss who received a low-sodium weight-reduction diet. Participants were enrolled at 38 academic and private practice sites in North America from March 2010 to November 2012 and followed up for 6 months (last visit in June 2013). All participants met the modified Dandy criteria for IIH and had a perimetric mean deviation (PMD) between -2 dB and -7 dB. The mean age was 29 years and all but 4 participants were women.InterventionsLow-sodium weight-reduction diet plus the maximally tolerated dosage of acetazolamide (up to 4 g/d) or matching placebo for 6 months.Main outcomes and measuresThe planned primary outcome variable was the change in PMD from baseline to month 6 in the most affected eye, as measured by Humphrey Field Analyzer. Perimetric mean deviation is a measure of global visual field loss (mean deviation from age-corrected normal values), with a range of 2 to -32 dB; larger negative values indicate greater vision loss. Secondary outcome variables included changes in papilledema grade, quality of life (Visual Function Questionnaire 25 [VFQ-25] and 36-Item Short Form Health Survey), headache disability, and weight at month 6.ResultsThe mean improvement in PMD was greater with acetazolamide (1.43 dB, from -3.53 dB at baseline to -2.10 dB at month 6; n = 86) than with placebo (0.71 dB, from -3.53 dB to -2.82 dB; n = 79); the difference was 0.71 dB (95% CI, 0 to 1.43 dB; P = .050). Mean improvements in papilledema grade (acetazolamide: -1.31, from 2.76 to 1.45; placebo: -0.61, from 2.76 to 2.15; treatment effect, -0.70; 95% CI, -0.99 to -0.41; P < .001) and vision-related quality of life as measured by the National Eye Institute VFQ-25 (acetazolamide: 8.33, from 82.97 to 91.30; placebo: 1.98, from 82.97 to 84.95; treatment effect, 6.35; 95% CI, 2.22 to 10.47; P = .003) and its 10-item neuro-ophthalmic supplement (acetazolamide: 9.82, from 75.45 to 85.27; placebo: 1.59, from 75.45 to 77.04; treatment effect, 8.23; 95% CI, 3.89 to 12.56; P < .001) were also observed with acetazolamide. Participants assigned to acetazolamide also experienced a reduction in weight (acetazolamide: -7.50 kg, from 107.72 kg to 100.22 kg; placebo: -3.45 kg, from 107.72 kg to 104.27 kg; treatment effect, -4.05 kg, 95% CI, -6.27 to -1.83 kg; P < .001).Conclusions and relevanceIn patients with IIH and mild visual loss, the use of acetazolamide with a low-sodium weight-reduction diet compared with diet alone resulted in modest improvement in visual field function. The clinical importance of this improvement remains to be determined.Trial registrationclinicaltrials.gov Identifier: NCT01003639.
| S-EPMC4362615 | biostudies-literature
Project description:The Idiopathic Intracranial Hypertension Treatment Trial (IIHTT) showed that acetazolamide provided a modest, significant improvement in mean deviation (MD). Here, we further analyze visual field changes over the 6-month study period.Of 165 subjects with mild visual loss in the IIHTT, 125 had perimetry at baseline and 6 months. We evaluated pointwise linear regression of visual sensitivity versus time to classify test locations in the worst MD (study) eye as improving or not; pointwise changes from baseline to month 6 in decibels; and clinical consensus of change from baseline to 6 months.The average study eye had 36 of 52 test locations with improving sensitivity over 6 months using pointwise linear regression, but differences between the acetazolamide and placebo groups were not significant. Pointwise results mostly improved in both treatment groups with the magnitude of the mean change within groups greatest and statistically significant around the blind spot and the nasal area, especially in the acetazolamide group. The consensus classification of visual field change from baseline to 6 months in the study eye yielded percentages (acetazolamide, placebo) of 7.2% and 17.5% worse, 35.1% and 31.7% with no change, and 56.1% and 50.8% improved; group differences were not statistically significant.In the IIHTT, compared to the placebo group, the acetazolamide group had a significant pointwise improvement in visual field function, particularly in the nasal and pericecal areas; the latter is likely due to reduction in blind spot size related to improvement in papilledema. (ClinicalTrials.gov number, NCT01003639.).
| S-EPMC4777274 | biostudies-literature
Project description:PurposeTo characterize visual field (VF) loss at the baseline visit and to evaluate VF quality control (QC) procedures in the Idiopathic Intracranial Hypertension Treatment Trial (IIHTT).MethodsThe Visual Field Reading Center (VFRC) evaluated 660 baseline VFs (1320 hemifields) from 165 enrolled patients. Three readers independently classified each superior and inferior hemifield and identified any abnormalities. A subset (20%) of the hemifields was reread to evaluate within- and between-reader agreements. The QC system addressed test parameters, patient data, and shipment errors.ResultsThe majority (60%) of the baseline hemifields consisted of localized nerve fiber bundle-type VF loss. Approximately one-third (31.5%) of all the classifications consisted of partial arcuate defects combined with an enlarged blind spot, making this the most common type of hemifield classification. Inferior hemifield loss was greater than superior loss for both study and nonstudy eyes. Reader agreements were >90% for both inferior and superior hemifields for two out of three readers. Test-retest reliability agreement for individual readers was 95% for both hemifields. There were few QC errors with only 5.48 error points per 100-point VF.ConclusionsThe most common type of IIHTT baseline hemifield abnormality was a localized nerve fiber bundle-like defect. Localized inferior hemifield loss was more common than superior hemifield loss. Quality control and within- and between-reader agreement were excellent for the IIHTT (ClinicalTrials.gov number, NCT01003639).
| S-EPMC4031894 | biostudies-literature
Project description:PurposeIdentifying and monitoring visual field (VF) defects due to optic neuritis (ON) relies on qualitative clinician interpretation. Archetypal analysis (AA), a form of unsupervised machine learning, is used to quantify VF defects in glaucoma. We hypothesized that AA can identify quantifiable, ON-specific patterns (as archetypes [ATs]) of VF loss that resemble known ON VF defects.MethodsWe applied AA to a dataset of 3892 VFs prospectively collected from 456 eyes in the Optic Neuritis Treatment Trial (ONTT), and decomposed each VF into component ATs (total weight = 100%). AA of 568 VFs from 61 control eyes was used to define a minimum meaningful (≤7%) AT weight and weight change. We correlated baseline ON AT weights with global VF indices, visual acuity, and contrast sensitivity. For eyes with a dominant AT (weight ≥50%), we compared the ONTT VF classification with the AT pattern.ResultsAA generated a set of 16 ATs containing patterns seen in the ONTT. These were distinct from control ATs. Baseline study eye VFs were decomposed into 2.9 ± 1.5 ATs. AT2, a global dysfunction pattern, had the highest mean weight at baseline (36%; 95% confidence interval, 33%-40%), and showed the strongest correlation with MD (r = -0.91; P < 0.001), visual acuity (r = 0.70; P < 0.001), and contrast sensitivity (r = -0.77; P < 0.001). Of 191 baseline VFs with a dominant AT, 81% matched the descriptive classifications.ConclusionsAA identifies and quantifies archetypal, ON-specific patterns of VF loss.Translational relevanceAA is a quantitative, objective method for demonstrating and monitoring change in regional VF deficits in ON.
| S-EPMC8787544 | biostudies-literature
Project description:To characterize the phenotype, headache-related disability, medical co-morbidities, use of symptomatic headache medications, and headache response to study interventions in the Idiopathic Intracranial Hypertension Treatment Trial (IIHTT).Patients with untreated IIH and mild vision loss enrolled in the IIHTT and randomized to acetazolamide (ACZ) and weight loss or placebo (PLB) and weight loss had prospective assessment of headache disability using the Headache Impact Test-6 (HIT-6) questionnaire. Subjects with headache at the baseline visit were assigned a headache phenotype using the International Classification for Headache Disorders version 3 beta (ICHD-3b). Medication overuse was determined using the participants' reported medication use for the preceding month and ICHD-3b thresholds for diagnosing medication overuse headache. We investigated relationships between headache disability and various other clinical characteristics at baseline and at 6 months.Headache was present in 139 (84%) of the 165 enrollees at baseline. The most common headache phenotypes were migraine (52%), tension-type headache (22%), probable migraine (16%), and probable tension-type headache (4%). Fifty-one (37%) participants overused symptomatic medications at baseline, most frequently simple analgesics. A similar amount of improvement in the adjusted mean (± standard error) HIT-6 score occurred in the ACZ (-9.56?±?1.05) and PLB groups (-9.11?±?1.14) at 6 months (group difference -0.45, 95% CI -3.50 to 2.60, P?=?.77). Headache disability did not correlate with any of the studies, variables of interest, which included: the lumbar puncture opening pressure at baseline or at 6 months, body mass index, the amount of weight lost, papilledema grade, perimetric mean deviation, or the use of hormonal contraception. Headache disability was significantly associated with patient-reported quality of life in the physical, mental, and visual domains.Headache was common, of varied character, disabling, and associated with poorer quality of life in our cohort of patients with mild visual impairment. The lack of correlation between headache disability and cerebrospinal fluid (CSF) pressure at baseline and at the end of the randomized phase of the study implies that headache in IIH may be related to factors other than intracranial hypertension, and that specific headache treatment is needed in addition to therapies directed at lowering CSF pressure.
| S-EPMC5799151 | biostudies-literature
Project description:Determine potential risk factors for progressive visual field loss in the Idiopathic Intracranial Hypertension Treatment Trial, a randomized placebo-controlled trial of acetazolamide in patients with idiopathic intracranial hypertension and mild visual loss concurrently receiving a low sodium, weight reduction diet.Logistic regression and classification tree analyses were used to evaluate potential risk factors for protocol-defined treatment failure (>2 dB perimetric mean deviation [PMD] change in patients with baseline PMD -2 to -3.5 dB or >3 dB PMD change with baseline PMD -3.5 to -7 dB).Seven participants (6 on diet plus placebo) met criteria for treatment failure. The odds ratio for patients with grades III to V papilledema vs those with grades I and II was 8.66 (95% confidence interval [CI] 1.65-?, p = 0.025). A 1-unit decrease in the number of letters correct on the ETDRS (Early Treatment Diabetic Retinopathy Study) chart at baseline was associated with an increase in the odds of treatment failure by a factor of 1.16 (95% CI 1.04-1.30, p = 0.005). Compared with female participants, the odds ratio for male participants was 26.21 (95% CI 1.61-433.00, p = 0.02). The odds of treatment failure were 10.59 times higher (95% CI 1.63-116.83, p = 0.010) for patients with >30 transient visual obscurations per month vs those with ?30 per month.Male patients, those with high-grade papilledema, and those with decreased visual acuity at baseline were more likely to experience treatment failure. All but one of these patients were treated with diet alone. These patients should be monitored closely and be considered for aggressive treatment of their idiopathic intracranial hypertension.
| S-EPMC4553022 | biostudies-literature
Project description:Treatment for a patient who presented with severe headaches and decreased vision caused by idiopathic intracranial hypertension was complicated by nonadherence and pregnancy, but the patient's symptoms resolved after a successful delivery.
| S-EPMC6364811 | biostudies-other
Project description:ObjectiveTo delineate the factors contributing to overdiagnosis of idiopathic intracranial hypertension (IIH) among patients seen in one neuro-ophthalmology service at a tertiary center.MethodsWe retrospectively reviewed new patients referred with a working diagnosis of IIH over 8 months. The Diagnosis Error Evaluation and Research taxonomy tool was applied to cases referred with a diagnosis of IIH and a discrepant final diagnosis.ResultsOf 1,249 patients, 165 (13.2%) were referred either with a preexisting diagnosis of IIH or to rule out IIH. Of the 86/165 patients (52.1%) with a preexisting diagnosis of IIH, 34/86 (39.5%) did not have IIH. The most common diagnostic error was inaccurate ophthalmoscopic examination in headache patients. Of 34 patients misdiagnosed as having IIH, 27 (27/34 [79.4%]; 27/86 [31.4%]) had at least one lumbar puncture, 29 (29/34 [85.3%]; 29/86 [33.7%]) had a brain MRI, and 8 (8/34 [23.5%]; 8/86 [9.3%]) had a magnetic resonance/CT venogram. Twenty-six had received medical treatment, 1 had a lumbar drain, and 4 were referred for surgery. In 8 patients (8/34 [23.5%]; 8/86 [9.3%]), an alternative diagnosis requiring further evaluation was identified.ConclusionsDiagnostic errors resulted in overdiagnosis of IIH in 39.5% of patients referred for presumed IIH, and prompted unnecessary tests, invasive procedures, and missed diagnoses. The most common errors were inaccurate ophthalmoscopic examination in headache patients and thinking biases, reinforcing the need for rapid access to specialists with experience in diagnosing optic nerve disorders. Indeed, the high prevalence of primary benign headaches and obesity in young women often leads to costly and invasive evaluations for presumed IIH.
| S-EPMC4776085 | biostudies-literature
Project description:Idiopathic intracranial hypertension (IIH) is a condition of unknown etiology often encountered in neurologic practice. It produces nonlocalizing symptoms and signs of raised intracranial pressure and, when left untreated, can result in severe irreversible visual loss. It most commonly occurs in obese women of childbearing age, but it can also occur in children, men, nonobese adults, and older adults. Although it is frequently associated with obesity, it can be associated with other conditions, such as obstructive sleep apnea and transverse cerebral venous sinus stenoses. Recent identification of subgroups at high risk for irreversible visual loss, including black patients, men, and patients with fulminant forms of IIH, help guide the optimal management and follow-up. Ongoing studies of venous anatomy and physiology in IIH patients, as well as a recently begun randomized clinical treatment trial, should provide further insight into this common yet poorly understood syndrome.
| S-EPMC3674489 | biostudies-other
Project description:ImportanceAlthough the central visual field (VF) in end-stage glaucoma may substantially vary among patients, structure-function studies and quality-of-life assessments are impeded by the lack of appropriate characterization of end-stage VF loss.ObjectiveTo provide a quantitative characterization and classification of central VF loss in end-stage glaucoma.Design, setting, and participantsThis retrospective cohort study collected data from 5 US glaucoma services from June 1, 1999, through October 1, 2014. A total of 2912 reliable 10-2 VFs of 1103 eyes from 1010 patients measured after end-stage 24-2 VFs with a mean deviation (MD) of -22 dB or less were included in the analysis. Data were analyzed from March 28, 2018, through May 23, 2019.Main outcomes and measuresCentral VF patterns were determined by an artificial intelligence algorithm termed archetypal analysis. Longitudinal analyses were performed to investigate whether the development of central VF defect mostly affects specific vulnerability zones.ResultsAmong the 1103 patients with the most recent VFs, mean (SD) age was 70.4 (14.3) years; mean (SD) 10-2 MD, -21.5 (5.6) dB. Fourteen central VF patterns were determined, including the most common temporal sparing patterns (304 [27.5%]), followed by mostly nasal loss (280 [25.4%]), hemifield loss (169 [15.3%]), central island (120 [10.9%]), total loss (91 [8.3%]), nearly intact field (56 [5.1%]), inferonasal quadrant sparing (42 [3.8%]), and nearly total loss (41 [3.7%]). Location-specific median total deviation analyses partitioned the central VF into a more vulnerable superonasal zone and a less vulnerable inferotemporal zone. At 1-year and 2-year follow-up, new defects mostly occurred in the more vulnerable zone. Initial encroachments on an intact central VF at follow-up were more likely to be from nasal loss (11 [18.4%]; P < .001). One of the nasal loss patterns had a substantial chance at 2-year follow-up (8 [11.0%]; P = .004) to shift to total loss, whereas others did not.Conclusions and relevanceIn this study, central VF loss in end-stage glaucoma was found to exhibit characteristic patterns that might be associated with different subtypes. Initial central VF loss is likely to be nasal loss, and 1 specific type of nasal loss is likely to develop into total loss.
| S-EPMC6990977 | biostudies-literature