Digital Health: Today's Solutions and Tomorrow's Impact.

Artificial intelligence (AI) is becoming an indispensable tool to augment decision making in different health care settings and by various members of the patient pathway, including the patient. AI provides the ability to optimize data to bring clinical decision support for clinicians and laboratorians and/or empower patients to actively participate in their own health care. Though there are many examples of AI in health care, the exact role of AI and digital health solutions is still taking shape. Although AI will not replace the clinician, those who do not adopt AI may in time, be left behind.

Real evidence to assess clinical testing interference risk (REACTIR): A strategy using real world data to assess the prevalence of interfering substances in patients undergoing clinical laboratory testing.

INTRODUCTION: Laboratory test interferences can cause spurious test results and patient harm. Knowing the frequency of various interfering substances in patient populations likely to be tested with a particular laboratory assay may inform test development, test utilization and strategies to mitigate interference risk. METHODS: We developed REACTIR (Real Evidence to Assess Clinical Testing Interference Risk), an approach using real world data to assess the prevalence of various interfering substances in patients tested with a particular type of assay. REACTIR uses administrative real world data to identify and subgroup patient cohorts tested with a particular laboratory test and evaluate interference risk. RESULTS: We demonstrate the application REACTIR to point of care (POC) blood glucose testing. We found that exposure to several substances with the potential to interfere in POC blood glucose tests, including N-acetyl cysteine (NAC) and high dose vitamin C was uncommon in most patients undergoing POC glucose tests with several key exceptions, such as burn patients receiving high dose IV-vitamin C or acetaminophen overdose patients receiving NAC. CONCLUSIONS: Findings from REACTIR may support risk mitigation strategies including targeted clinician education and clinical decision support. Likewise, adaptations of REACTIR to premarket assay development may inform optimal assay design and assessment.

Development of a "meta-model" to address missing data, predict patient-specific cancer survival and provide a foundation for clinical decision support.

OBJECTIVE: Like most real-world data, electronic health record (EHR)-derived data from oncology patients typically exhibits wide interpatient variability in terms of available data elements. This interpatient variability leads to missing data and can present critical challenges in developing and implementing predictive models to underlie clinical decision support for patient-specific oncology care. Here, we sought to develop a novel ensemble approach to addressing missing data that we term the "meta-model" and apply the meta-model to patient-specific cancer prognosis. MATERIALS AND METHODS: Using real-world data, we developed a suite of individual random survival forest models to predict survival in patients with advanced lung cancer, colorectal cancer, and breast cancer. Individual models varied by the predictor data used. We combined models for each cancer type into a meta-model that predicted survival for each patient using a weighted mean of the individual models for which the patient had all requisite predictors. RESULTS: The meta-model significantly outperformed many of the individual models and performed similarly to the best performing individual models. Comparisons of the meta-model to a more traditional imputation-based method of addressing missing data supported the meta-model's utility. CONCLUSIONS: We developed a novel machine learning-based strategy to underlie clinical decision support and predict survival in cancer patients, despite missing data. The meta-model may more generally provide a tool for addressing missing data across a variety of clinical prediction problems. Moreover, the meta-model may address other challenges in clinical predictive modeling including model extensibility and integration of predictive algorithms trained across different institutions and datasets.

Multi-Site PCR-based CMV viral load assessment-assays demonstrate linearity and precision, but lack numeric standardization: a report of the association for molecular pathology.

Viralload (VL) assessment of cytomegalovirus (CMV) by real-time PCR is an important tool for diagnosing and monitoring CMV viremia in patients with compromised immune systems. We report results from a sample exchange organized by members of the Association for Molecular Pathology that compared PCR results from 23 laboratories; 22 such laboratories used a laboratory-developed real-time PCR assay and one laboratory used a competitive PCR assay. The samples sent to each laboratory were comprised of a dilution panel of CMV virion-derived reference materials that ranged from 0 to 500,000 copies/ml. Accuracy, linearity, and intralaboratory precision were established for the different laboratory-developed assays. Overall, PCR results were linear for each laboratory (R(2) > 0.97 in all but two). While 13 laboratories showed no significant quantitative assay bias, 10 laboratories reported VLs that were significantly different compared with expected values (bias range, -0.82 to 1.4 logs). The intralaboratory precision [mean coefficient of variance of 2% to 5% (log-scale)] suggested that changes in VLs of less than 3- to fivefold may not be significantly different. There was no significant association between laboratory-specific technical variables (PCR platform, calibrator, extraction method) and assay linearity or accuracy. These data suggested that, within each laboratory, relative VL values were linear, but additional method standardization and a CMV DNA reference standard are needed to allow laboratories to achieve comparable numeric results.

Detection of an apparent homozygous 3120G>A cystic fibrosis mutation on a routine carrier screen.

A 28-year-old Caucasian female with no personal or family history of cystic fibrosis (CF) presented for preconception counseling and screening. Cystic fibrosis transmembrane conductance regulator (CFTR) mutation analysis using the Inno-LiPa CFTR assay revealed lack of hybridization for both the wild-type and mutant oligonucleotides for 3120+1G>A. This region was sequenced, and an apparent homozygous 3120G>A mutation was detected. Additional testing revealed an abnormal sweat chloride (77 mmol/L). Review of systems was essentially unremarkable with an absence of sinus symptoms, occasional nonproductive cough, and no features of malabsorption. Physical examination, chest X-ray, and pulmonary function tests were within normal limits. Only two other patients (siblings) with homozygous 3120G>A mutations have been reported (http://www.genet.sickkids.on.ca/cftr/). Both siblings had pancreatic insufficiency, mild pulmonary symptoms, and abnormal sweat chloride levels. Our findings suggest that a homozygous mutation of a G>A conversion at 3120 is associated with abnormal CFTR function and either a mild form of CF or no overt symptoms of disease, emphasizing the difficulties in assigning genotype/phenotype correlation.

Dystroglycan expression in the mouse cochlea.

Viable dominant spotting (W(v)/W(v)) mice have a c-kit gene mutation, which impedes the migration of neural crest cells to the developing cochlea where they normally differentiate into intermediate cells (ICs). A prominent pathological feature shared by these mutants and the aging human and gerbil cochlea is thickening of the basement membrane (BM) of strial capillaries. Atrophy of strial capillaries in the aging gerbil has been associated with changes in the expression of dystroglycan (DG), a cell-surface receptor that regulates BM assembly. Here we evaluated the expression of DG in W(v)/W(v) mutant and C57BL/6J wild-type mice to investigate the possible role of ICs in regulating strial capillary BM homeostasis. The DG gene product was identified in lateral wall dissections from both W(v)/W(v) mutant and wild-type mice by reverse transcription-polymerase chain reaction. Subunit-specific antibodies were employed to localize the alpha and beta subunits of the DG heterodimer. Some sites in both wild-type and mutant mice, such as the subepithelial BM lining the scala media and regions of contact between selected epithelial cells, expressed alpha-DG alone. Other sites such as the perineural BM and the perivascular BM subtending strial capillaries and capillaries in the central portion of the auditory nerve coexpressed alpha- and beta-DG. The strong diffuse staining for alpha-DG along the basolateral membrane of strial marginal cells disappeared with advancing strial degeneration in abnormal turns of W(v)/W(v) mutants. Variations in staining intensity for both alpha- and beta-DG also occurred in the subendothelial BM of strial capillaries in turns lacking ICs and appeared to correspond with the degree of capillary atrophy. The results support the possibility that ICs play a role in the homeostasis of the strial capillary BM.

Dystroglycan expression in the developing and senescent gerbil cochlea.

Dystroglycan (DG) forms part of a cell surface laminin receptor complex and is believed to play a critical role in the assembly and homeostasis of basement membranes (BM). The receptor complex is made up of alpha- and beta-DG subunits and is found in muscle, epithelial and nerve tissue. In the cochlea, DG may be involved in the abnormal accumulation of laminin seen in the thickened BM of strial capillaries with age. This excess deposition of laminin is thought to lead to capillary necrosis and contribute to degeneration of the stria vascularis (SV). Here we assessed the presence and distribution of DG in the developing, mature and senescent gerbil cochlea in order to ascertain whether altered patterns of expression are a factor in age-related pathology. Western blots of proteins isolated from the entire cochlea demonstrated the presence of the alpha-DG subunit. mRNA encoding DG was identified in microdissected specimens of the lateral wall and the combined organ of Corti/modiolus by RT-PCR analysis. Immunohistochemical experiments localized alpha-DG in epithelial BMs and regions of epithelial cell-cell contact with no intervening BM in the developing and mature cochlea. Immunoreactive alpha-DG was present in the BM underlying strial capillaries and in vessels of the central portion of the auditory nerve, but was not detected in any other vessels in the cochlea. Age-related changes in alpha-DG expression were observed only in the SV where a marked decrease in alpha-DG immunoreactivity was seen in the BM of strial capillaries as well as throughout the SV. The results demonstrate the selective expression of alpha-DG in both BM and non-BM sites in the mature cochlea and suggests its involvement in both developmental and aging processes.