Biomedical subjects

Michael F Chiang

Publications and source records attributed to Michael F Chiang.

13 recordsLinked to original sources

PubMed2006

Representation of ophthalmology concepts by electronic systems: intercoder agreement among physicians using controlled terminologies.

OBJECTIVE: To assess intercoder agreement for ophthalmology concepts by 3 physician coders using 5 controlled terminologies (International Classification of Diseases 9, Clinical Modification [ICD9CM]; Current Procedural Terminology, fourth edition; Logical Observation Identifiers, Names, and Codes [LOINC]; Systematized Nomenclature of Medicine, Clinical Terms [SNOMED-CT]; and Medical Entities Dictionary). DESIGN: Noncomparative case series. PARTICIPANTS: Five complete ophthalmology case presentations selected from a publicly available journal. METHODS: Each case was parsed into discrete concepts. Electronic or paper browsers were used independently by 3 physician coders to assign a code for every concept in each terminology. A match score representing adequacy of assignment for each concept was assigned on a 3-point scale (0, no match; 1, partial match; 2, complete match). For every concept, the level of intercoder agreement was determined by 2 methods: (1) based on exact code matching with assignment of complete agreement when all coders assigned the same code, partial agreement when 2 coders assigned the same code, and no agreement when all coders assigned different codes, and (2) based on manual review for semantic equivalence of all assigned codes by an independent ophthalmologist to classify intercoder agreement for each concept as complete agreement, partial agreement, or no agreement. Subsequently, intercoder agreement was calculated in the same manner for the subset of concepts judged to have adequate coverage by each terminology, based on receiving a match score of 2 by at least 2 of the 3 coders. MAIN OUTCOME MEASURES: Intercoder agreement in each controlled terminology: complete, partial, or none. RESULTS: Cases were parsed into 242 unique concepts. When all concepts were analyzed by manual review, the proportion of complete intercoder agreement ranged from 12% (LOINC) to 44% (SNOMED-CT), and the difference in intercoder agreement between LOINC and all other terminologies was statistically significant (P<0.004). When only concepts with adequate terminology were analyzed by manual review, the proportion of complete intercoder agreement ranged from 33% (LOINC) to 64% (ICD9CM), and there were no statistically significant differences in intercoder agreement among any pairs of terminologies. CONCLUSIONS: The level of intercoder agreement for ophthalmic concepts in existing controlled medical terminologies is imperfect. Intercoder reproducibility is essential for accurate and consistent electronic representation of medical data.

Decision Support Systems, Clinical↗

PubMed2006

Accuracy and reliability of remote retinopathy of prematurity diagnosis.

OBJECTIVE: To determine the accuracy and reliability of retinopathy of prematurity (ROP) diagnosis using remote review of digital images by 3 masked ophthalmologist readers. METHODS: An atlas was compiled of 410 retinal photographs from 163 eyes of 64 low-birth-weight infants taken using a wide-angle digital fundus camera. All the images were independently reviewed by 3 readers, and the diagnosis in each eye was classified into 1 of 4 ordinal categories: no ROP, mild ROP, type 2 prethreshold ROP, or ROP requiring treatment. Findings were compared with a reference standard of dilated indirect ophthalmoscopy with scleral depression performed by an experienced pediatric ophthalmologist. RESULTS: Sensitivities/specificities of the diagnosis of any ROP were 0.845/0.910 for the first reader, 0.816/0.955 for the second reader, and 0.864/0.493 for the third reader. Sensitivities/specificities of the diagnosis of ROP requiring treatment were 0.850/0.960 for the first reader, 0.850/0.973 for the second reader, and 0.900/0.953 for the third reader. When ROP was classified into ordinal categories, the overall weighted kappa for interreader reliability was 0.743. Intrareader reliability for detection of low-risk prethreshold ROP or worse was 100% for all readers. CONCLUSION: The accuracy, interreader reliability, and intrareader reliability of remote diagnosis of clinically relevant ROP based on digital imaging are substantial.

Birth Weight↗

PubMed2006

Retinal nerve fiber layer thickness in normal children measured with optical coherence tomography.

PURPOSE: To measure the peripapillary retinal nerve fiber layer (RNFL) thickness in normal children. DESIGN: Observational cross-sectional study. PARTICIPANTS: Ninety-two eyes of 92 normal children ages 4 to 17 years presenting to the Ophthalmology Clinic at the Harkness Eye Institute, Department of Ophthalmology, Columbia University. METHODS: Retinal nerve fiber layer thickness was measured with optical coherence tomography (OCT). Patient cooperation and signal strength of the OCT scans were assessed. Optic disc photographs were evaluated by a glaucoma specialist in a masked fashion. Eyes with abnormal optic discs were excluded. One eye of each subject was randomly selected for statistical analysis. The effect of several factors on RNFL thickness was investigated statistically. MAIN OUTCOME MEASURES: Retinal nerve fiber layer thickness. RESULTS: Ninety-one percent of the study subjects were Hispanic; 8%, African American; and 1%, Caucasian. Optical coherence tomography measurements were obtained in 117 of 121 (96.7%) subjects, and disc photographs were available for 92 of them. Mean age (+/-standard deviation [SD]) was 9.7+/-2.7 years. Mean global RNFL thickness (+/-SD) was 107.0+/-11.1 microm (range, 78.1-134.6). The RNFL was thickest inferiorly (136.9+/-16.9 microm) and superiorly (135.4+/-19.3 microm), thinner nasally (83.0+/-18.0 microm), and thinnest temporally (72.5+/-13.4 microm). In univariate regression analysis, age (P = 0.013) and refraction (P<0.001) had a significant effect on RNFL thickness; age had a significant effect on refraction (P<0.001). When controlling for refraction, age no longer had a significant effect. CONCLUSIONS: Optical coherence tomography can be used to measure RNFL thickness in children. Refraction had an effect on RNFL thickness. In normal children, variation in RNFL thickness is large. The normative data provided by this study may assist in identifying changes in RNFL thickness in children.

Adolescent↗

PubMed2005

Representation of ophthalmology concepts by electronic systems: adequacy of controlled medical terminologies.

OBJECTIVE: To assess the adequacy of 5 controlled medical terminologies (International Classification of Diseases 9, Clinical Modification [ICD9-CM]; Current Procedural Terminology 4 [CPT-4]; Systematized Nomenclature of Medicine, Clinical Terms [SNOMED-CT]; Logical Identifiers, Names, and Codes [LOINC]; Medical Entities Dictionary [MED]) for representing concepts in ophthalmology. DESIGN: Noncomparative case series. PARTICIPANTS: Twenty complete ophthalmology case presentations were sequentially selected from a publicly available ophthalmology journal. METHODS: Each of the 20 cases was parsed into discrete concepts, and each concept was classified along 2 axes: (1) diagnosis, finding, or procedure and (2) ophthalmic or medical concept. Electronic or paper browsers were used to assign a code for every concept in each of the 5 terminologies. Adequacy of assignment for each concept was scored on a 3-point scale. Findings from all 20 case presentations were combined and compared based on a coverage score, which was the average score for all concepts in that terminology. MAIN OUTCOME MEASURES: Adequacy of assignment for concepts in each terminology, based on a 3-point Likert scale (0, no match; 1, partial match; 2, complete match). RESULTS: Cases were parsed into 1603 concepts. SNOMED-CT had the highest mean overall coverage score (1.625+/-0.667), followed by MED (0.974+/-0.764), LOINC (0.781+/-0.929), ICD9-CM (0.280+/-0.619), and CPT-4 (0.082+/-0.337). SNOMED-CT also had higher coverage scores than any of the other terminologies for concepts in the diagnosis, finding, and procedure categories. Average coverage scores for ophthalmic concepts were lower than those for medical concepts. CONCLUSIONS: Controlled terminologies are required for electronic representation of ophthalmology data. SNOMED-CT had significantly higher content coverage than any other terminology in this study.

Humans↗

PubMed2005

Mining a clinical data warehouse to discover disease-finding associations using co-occurrence statistics.

This paper applies co-occurrence statistics to discover disease-finding associations in a clinical data warehouse. We used two methods, chi2 statistics and the proportion confidence interval (PCI) method, to measure the dependence of pairs of diseases and findings, and then used heuristic cutoff values for association selection. An intrinsic evaluation showed that 94 percent of disease-finding associations obtained by chi2 statistics and 76.8 percent obtained by the PCI method were true associations. The selected associations were used to construct knowledge bases of disease-finding relations (KB-chi2, KB-PCI). An extrinsic evaluation showed that both KB-chi2 and KB-PCI could assist in eliminating clinically non-informative and redundant findings from problem lists generated by our automated problem list summarization system.

Chi-Square Distribution↗

PubMed2005

Assessment of image-based technology: impact of referral cutoff on accuracy and reliability of remote retinopathy of prematurity diagnosis.

Telemedicine has potential to improve the delivery, quality, and accessibility of ophthalmic care for infants with Retinopathy of Prematurity (ROP). Using a telemedicine screening strategy, three potential diagnostic cutoffs may be used to define disease that warrants ophthalmologic referral: presence of any ROP, presence of moderate ("type-2 prethreshold") ROP, or presence of severe ROP requiring treatment. This study examines the relationship between accuracy and reliability of diagnosis by three masked ophthalmologist graders at each of these diagnostic cutoffs. The sensitivity, specificity, inter-grader reliability, and intra-grader reliability showed significant variation depending on the diagnostic cutoff, with best results at cutoffs of type-2 prethreshold ROP or treatment-requiring ROP. Before the large-scale adoption of telemedicine for image-based screening of diseases such as ROP, standards defining clinically-relevant referral cutoffs must be established, and diagnostic accuracy and reliability at these cutoffs must be characterized.

Atlases as Topic↗

PubMed2004

Incidence of retinopathy of prematurity from 1996 to 2000: analysis of a comprehensive New York state patient database.

OBJECTIVE: To determine the current incidence of retinopathy of prematurity (ROP) in New York state. DESIGN: Population-based cohort study. PARTICIPANTS: Newborn infants (15 691) with initial hospital length of stay >28 days and date of discharge from January 1, 1996, to December 31, 2000. Subjects were identified from the New York Statewide Planning and Research Cooperative System (SPARCS) database, which contains information about every patient hospitalized in New York during this period. METHODS: Demographic and clinical information about eligible infants was abstracted by searching the SPARCS database. Patients with a discharge diagnosis of ROP or who underwent laser retinal photocoagulation, scleral buckle, or pars plana vitrectomy were identified by searching for appropriate discharge and procedure codes. Incidence of ROP in the study population was determined and analyzed on the basis of birth weight. MAIN OUTCOME MEASURES: Incidence of any ROP, laser photocoagulation, scleral buckle, and pars plana vitrectomy in study population. RESULTS: On the basis of SPARCS coding, the overall incidence of any ROP among all newborn infants in New York state during the study period was 0.2% (2284 of 1 167 427), or 1 in 511. The incidence of any ROP in the study population of newborns with initial hospital length of stay >28 days was 20.3% (2152 of 10 596) among infants with birth weight <1500 g and 27.3% (1839 of 6745) among infants with birth weight <1200 g. Among study patients with any ROP, the proportion who underwent laser photocoagulation during initial hospital stay was 9.5% (218 of 2284), and the proportion who underwent scleral buckle or vitrectomy surgery was 0.5% (12 of 2284). Seventeen study newborns with birth weight > or =2000 g had a discharge diagnosis of ROP, although none of these patients required laser or incisional surgery during hospitalization. CONCLUSIONS: This study involves the largest known cohort of newborns that has been analyzed for ROP. The incidence of ROP in this study is lower than results from previous multicenter clinical trials. However, the diagnosis of ROP in 17 study newborns with birth weight > or =2000 g deserves further investigation and may have implications for ROP screening protocols.

Birth Weight↗

PubMed2004

Evaluation of consumer health website accessibility by users with sensory and physical disabilities.

Growth of the World Wide Web is beginning to create new opportunities for direct patient access to health care resources. At the same time, advances in medical care have produced demographic shifts in which an increasing number of patients have sensory and physical disabilities that may limit their ability to access these new information tools. This study uses automated and manual methods to measure the compliance of 30 popular consumer health websites with the World Wide Web Consortium (W3C) content accessibility guidelines. Among these 30 sites, 22 failed to satisfy at least one W3C Priority-1 accessibility checkpoint, making it impossible for some groups of disabled users to access information from them. All websites failed to satisfy at least one Priority-2 checkpoint, making it difficult for some groups to access information from the sites. These results suggest that accessibility of many consumer health websites to disabled users is very limited. The health informatics community must become more aware of this problem, particularly because many critical accessibility problems may be easily addressed if they are recognized

Persons with Disabilities↗

PubMed2003

An experimental system for comparing speed, accuracy, and completeness of physician data entry using electronic and paper methods.

Electronic medical record (EMR) systems have important potential advantages over traditional paper-based systems, but they require that physicians assume responsibility for data entry. However, little is known about the quality of physician data entry in electronic systems. This study describes a system for comparing the speed, accuracy, and completeness of examination data entry using electronic and paper methods. Data will be shown to demonstrate that this may be a simple, reproducible, and useful technique.

Humans↗

PubMed2002

Software engineering risk factors in the implementation of a small electronic medical record system: the problem of scalability.

The successful implementation of clinical information systems is difficult. In examining the reasons and potential solutions for this problem, the medical informatics community may benefit from the lessons of a rich body of software engineering and management literature about the failure of software projects. Based on previous studies, we present a conceptual framework for understanding the risk factors associated with large-scale projects. However, the vast majority of existing literature is based on large, enterprise-wide systems, and it unclear whether those results may be scaled down and applied to smaller projects such as departmental medical information systems. To examine this issue, we discuss the case study of a delayed electronic medical record implementation project in a small specialty practice at Columbia-Presbyterian Medical Center. While the factors contributing to the delay of this small project share some attributes with those found in larger organizations, there are important differences. The significance of these differences for groups implementing small medical information systems is discussed.

Computer Systems↗

PubMedDate not supplied

Computer and World Wide Web accessibility by visually disabled patients: problems and solutions.

Rapid advances in information technology have dramatically transformed the world during the past several decades. Access to computers and the World Wide Web is increasingly required for education and employment, as well as for many activities of daily living. Although these changes have improved society in many respects, they present an obstacle for visually disabled patients who may have significant difficulty processing the visual cues presented by modern graphical user interfaces. This article reviews the specific barriers to computer and Web access faced by visually disabled patients, describes clinical evaluation methods, summarizes traditional low vision methods as well as newer assistive computer technologies for universal accessibility, and discusses emerging technologies and future directions in this area.

Access to Information↗

PubMedDate not supplied

Comparison of ultrasonic and ophthalmoscopic evaluation of retinopathy of prematurity.

PURPOSE: Screening for detection of retinopathy of prematurity (ROP) currently is limited to indirect ophthalmoscopy, which requires considerable examiner skill and experience. We investigated whether conventional 10 MHz B-scan ultrasonography could document the clinical stages of ROP as accurately as indirect ophthalmoscopy. METHODS: Thirty-four eyes of 18 neonates were examined by masked, independent observers with indirect ophthalmoscopy and digitally recorded 10-MHz B-scan ultrasonography. After pupil dilation and lid speculum placement, the retinologist recorded the stage of retinopathy with a retinal drawing. The ultrasonographer, without use of papillary mydriatics or lid speculum, determined the presence or absence of a ridge or tractional elements, if present on the ridge. RESULTS: Ultrasound grade correlated with clinical grade (R = .79, P < .001). However, nine eyes were overdiagnosed by one stage, and one eye, in which a peripheral detachment was mistaken for an artifact, was underdiagnosed. CONCLUSIONS: Ten-megahertz ultrasonography offers the potential of imaging and detecting the clinical stages of ROP; the use of higher ultrasound frequencies, now becoming commercially available, is likely to enhance diagnostic accuracy. Care must be taken to distinguish between artifact and true anatomical structures in noncontact ultrasound examinations. Neonates with suspected ROP could be screened with B-scan ultrasonography by neonatal personnel without pupillary dilatation or lid speculum, thus eliminating potential morbidity, and clinically significant cases of ROP then could be referred to the retinologist.

Fundus Oculi↗

PubMedDate not supplied

Is there a role for high-frequency ultrasonography in clinical staging of retinopathy of prematurity?

PURPOSE: To compare 20-MHz ultrasonography with ophthalmoscopy in the staging of retinopathy of prematurity (ROP). METHODS: We used a handheld 20-MHz ultrasound system to examine 38 eyes of 19 neonates who had an indirect ophthalmoscopic examination (diagnosis masked) within 48 hours prior to ultrasonography. Determination of ROP by indirect ophthalmoscopy was compared with independent identification of ultrasonic features indicative of ROP stages. RESULTS: In masked cases, ultrasonography correctly identified 13 of 18 (72.2% sensitivity) eyes with zone 2, stage 2 ROP or worse and 19 of 20 (95.0% specificity) normal eyes (zone 2, stage 1 ROP or better) (chi-square, 18.4; P < .001). CONCLUSION: In the neonatal nursery, a 20-MHz handheld ultrasound system may be an effective screening tool to initially identify ROP for further ophthalmoscopic evaluation and management.

Anterior Eye Segment↗