Tao brush endometrial cytology is a sensitive diagnostic tool for cancer and hyperplasia among women presenting to clinic with abnormal uterine bleeding.

BACKGROUND: Abnormal uterine bleeding requires the investigation of the endometrium. Histology is typically used but there remains room for the improvement and use of cytology. METHODS: Women presenting for clinically indicated office endometrial biopsy were prospectively enrolled. Tao endometrial brushing and office endometrial biopsy were performed, and surgical procedure if clinically indicated. Tao brush cytology specimens were blindly reviewed by up to three pathologists, consensus obtained, and scored as: benign, atypical (favor benign), suspicious, positive for malignancy, or non-diagnostic. Cytology and histology were compared to surgical pathology to determine sensitivity, specificity, positive, and negative predictive values to detect AH (atypical hyperplasia) or EC (endometrial cancer). RESULTS: Clinical indications of 197 enrolled patients included postmenopausal bleeding (90, 45.7%), abnormal uterine bleeding (94, 47.7%), and abnormal endometrium on ultrasound without bleeding (13, 6.6%). Of the 197 patients, 185 (93.9%) had cytology score consensus and a total of 196 (99.5%) had consensus regarding cytology positivity. Surgical pathology diagnoses (N = 85) were 13 (15.3%) FIGO grade 1 or 2 EC, 3 (3.5%) AH, and 69 (81.2%) benign endometrium. Sensitivity and specificity to detect EC or AH were 93.7% and 100%, respectively, via endometrial biopsy; 87.5% and 63.8%, respectively, via endometrial cytology when scores of malignancy, suspicious, or atypical were considered positive. CONCLUSIONS: In a high-risk population, Tao brush endometrial cytology showed high sensitivity to detect AH and EC comparable to biopsy histology when considering scores of malignancy, suspicious, atypical, and non-diagnostic. Revisiting the potential value of endometrial cytology in the contemporary era of endometrial diagnostic workup is warranted.

Quantifying procedural pain associated with office gynecologic tract sampling methods.

OBJECTIVE: Emerging technologies may enable detection of endometrial cancer with methods that are less invasive than standard biopsy methods. This study compares patient pain scores among 3 office gynecologic tract sampling methods and explores their potential determinants. METHODS: A prospective study including 3 sampling methods (tampon, Tao brush (TB), endometrial biopsy (EB)) was conducted between December 2015 and August 2017 and included women ≥45 years of age presenting with abnormal uterine bleeding, postmenopausal bleeding, or thickened endometrial stripe. Patients rated pain after each sampling procedure using a 100-point visual analog scale (VAS). RESULTS: Of 428 enrolled, 190 (44.39%) patients underwent all 3 sampling methods and reported a VAS score for each. Nearly half were postmenopausal (n = 93, 48.9%); the majority were parous (172, 90.5%) of which 87.8% had at least one vaginal delivery. Among the 190 patients, the median (IQR) pain score was significantly lower for sampling via tampon (0 [0,2]) compared to TB (28 [12, 52]) or EB (32 [15, 60]) (both p < 0.001, Wilcoxon signed rank test). Among women who underwent tampon sampling, age and pain scores showed a weak positive correlation (Spearman rank correlation, r = 0.14; p = 0.006); EB sampling was associated with a weak inverse correlation between parity and pain scores (r = -0.14; p = 0.016). CONCLUSION: Gynecologic tract sampling using a tampon had significantly lower pain than both EB and TB. Pain with tampon sampling was positively correlated with age and pain with EB sampling was inversely correlated with parity. Pain scores for TB and EB were not significantly related to age, menopausal status, or BMI.

Building a smart FNA cart: When Google meets cytology.

BACKGROUND: The appropriate management of a fine needle aspiration (FNA) supply cart and equipment set up is essential to ensure the smooth and optimal operation of a busy FNA clinic. We applied Lean strategies such as value stream mapping (VSM), the 5S method (Sort, Set in order, Shine, Standardize, Sustain), and Kanban to remove waste and improve patient flow in an FNA clinic. METHODS: The workflow analysis suggested that existent problems such as suboptimal inventory management and unavailability of standard operating procedures (SOPs) caused a 10% to 85% increase in total procedure time. To improve inventory management, we created a 2-bin Kanban system. We used the "Scan to Web" app and a Google Drive form to create a cost-effective electronic inventory management system. We distributed the essential SOPs in the format of video clips using our YouTube channel and leveraged barcode technology to access the links. RESULTS: Upon completion of our process improvement project, we succeeded to eliminate the stock-out events and maintain a process cycle efficiency of 87%. The 5S audit checklist result increased from 6% to 100% implementation, which is consistent with focused improvement. The developed inventory system enabled us to track the supply usage, forecast demands, and improve the accuracy of orders. CONCLUSIONS: Lean methods such as VSM, 5S, and Kanban combined with open source technologies can be implemented to ensure material availability, track inventory, and provide immediate access to SOPs on demand. The developed system also led to increased efficiency and improved flow, as well as responsiveness to changes in demand.

Investigation of Atyp.C using UF-5000 flow cytometer in patients with a suspected diagnosis of urothelial carcinoma: a single-center study.

BACKGROUND: This study evaluated the predictive power of Atyp.C (a parameter of UF-5000 flow cytometer) for patients with a suspected diagnosis of urothelial carcinoma. METHODS: We analyzed 163 urine specimens from 128 patients with suspected urothelial carcinoma using a fully automated fluorescence flow cytometry analyzer (UF-5000) and evaluated its performance on identifying atypical/malignant urothelial cells. From January 1, 2019 to April 4, 2019, all consecutive specimens for urinary cytopathology were enrolled. RESULTS: Of the specimens with urinary cytopathology, 67 specimens (41.1%) revealed abnormal findings in cytology analysis. Among them, 20 specimens (12.3%) were diagnosed as atypical urothelial cells, 26 specimens (16.0%) as suspicious for malignancy (S-malignancy), and 21 specimens (12.9%) as confirmed malignancy. The UF-5000 findings were positive in 59 specimens (36.2%); therefore, the agreement with cytopathology was 73.0%. Using follow-up histologic diagnosis of urothelial carcinoma with or without urinary tract cytology (UTCy) as a reference standard (suspicious and confirmed malignancy were the positive criteria for UTCy), the sensitivity was 59.0%, specificity was 82.1%, positive predictive value was 75.0%, negative predictive value was 68.8%, and the agreement was 71.1%. CONCLUSIONS: It is worth knowing and reporting that the Atyp.C assay may be used as an accessory test for patients with suspected urothelial carcinoma, based on its ability to identify high-risk patients who might need closer follow-up or additional medical treatment.

Two-site evaluation of a new workflow for the detection of malignant cells on the Sysmex XN-1000 body fluid analyzer.

INTRODUCTION: The presence of high fluorescent cells (HF-BF) on the Sysmex XN-1000 hematology analyzers has gained interest regarding the prediction of malignant cells in body fluids, but lacks sensitivity. We aimed to increase this sensitivity by combining HF-BF value, automated results, and clinical information. METHODS: We evaluated a new workflow for the management of body fluids in the hematology laboratory, including the HF-BF criterion and clinical information. In two laboratories, 1623 serous fluids were retrospectively analyzed on the XN-1000 BF mode. All samples were morphologically screened for malignant cells. Optimal HF-BF cutoffs were determined to predict their presence. Thereafter, the added value of clinical information was evaluated. Other reflex testing rules (eosinophilic count >5% and presence of the WBC Abnormal Scattergram flag) were also used to refine our workflow. RESULTS: Optimal HF-BF cutoffs in the two hematology centers were 108 and 45 cells/µL, yielding a sensitivity/specificity of 66.7/93.6% and 86.8/66.6% for malignant cell detection. When adding clinical information, sensitivity/specificity evolved to 100.0/68.9% and 100.0%/not determined. Of 104 samples containing malignant cells, 97 had positive clinical information; the remainder had a HF-BF > cutoff. CONCLUSION: Combining clinical information and HF-BF reached 100% sensitivity for malignant cell detection in body fluid analysis. Lack of robustness of the optimal HF-BF cutoff deserves the use of local cutoffs. Rapid automated results reporting from the XN-1000 BF mode are also feasible in clinical practice. Prospective evaluation of the workflow is needed before its implementation in clinical practice.

Point-of-care oral cytology tool for the screening and assessment of potentially malignant oral lesions.

BACKGROUND: The effective detection and monitoring of potentially malignant oral lesions (PMOL) are critical to identifying early-stage cancer and improving outcomes. In the current study, the authors described cytopathology tools, including machine learning algorithms, clinical algorithms, and test reports developed to assist pathologists and clinicians with PMOL evaluation. METHODS: Data were acquired from a multisite clinical validation study of 999 subjects with PMOLs and oral squamous cell carcinoma (OSCC) using a cytology-on-a-chip approach. A machine learning model was trained to recognize and quantify the distributions of 4 cell phenotypes. A least absolute shrinkage and selection operator (lasso) logistic regression model was trained to distinguish PMOLs and cancer across a spectrum of histopathologic diagnoses ranging from benign, to increasing grades of oral epithelial dysplasia (OED), to OSCC using demographics, lesion characteristics, and cell phenotypes. Cytopathology software was developed to assist pathologists in reviewing brush cytology test results, including high-content cell analyses, data visualization tools, and results reporting. RESULTS: Cell phenotypes were determined accurately through an automated cytological assay and machine learning approach (99.3% accuracy). Significant differences in cell phenotype distributions across diagnostic categories were found in 3 phenotypes (type 1 ["mature squamous"], type 2 ["small round"], and type 3 ["leukocytes"]). The clinical algorithms resulted in acceptable performance characteristics (area under the curve of 0.81 for benign vs mild dysplasia and 0.95 for benign vs malignancy). CONCLUSIONS: These new cytopathology tools represent a practical solution for rapid PMOL assessment, with the potential to facilitate screening and longitudinal monitoring in primary, secondary, and tertiary clinical care settings.

Assessment of clinical performance of an ultrasensitive nanowire assay for detecting human papillomavirus DNA in urine.

OBJECTIVE: To evaluate whether HPV DNA in urine has potential advantages as an alternative biomarker for HPV-based cervical cancer screening. METHODS: Among patients with Cobas HPV test results, a total of 67 HPV-positive (n = 42) and -negative (n = 25) women who agreed to participate in this study were willing to provide paired cervical and urine samples, and we observed concordance between sample types from each patient in identifying HPV genotypes using the nanowire assay. RESULTS: We detected high-risk strains of HPV DNA in unprocessed urine specimens using polyethyleneimine-conjugated nanowires (PEI-NWs). Concordance for high-risk HPV (hrHPV) between paired urine and cervical samples was 90.4% (κ = 0.90; 95% CI: 0.80-100.00). The virological sensitivity and specificity for detection of HPV DNA from a small urine sample (200 µL) were 81.3% (κ = 0.83; 95% CI: 62.1-100.0) and 98.0% (κ = 0.83; 95% CI: 94.2-100.0) for HPV16 group, 100.0% (κ = 0.65; 95% CI: 100.0-100.0) and 95.3% (κ = 0.65; 95% CI: 90.1-100.0) for HPV18 group, and 96.4% (κ = 0.97; 95% CI: 89.6-100.0) and 100.0% (κ = 0.97; 95% CI: 100.0-100.0) for other hrHPV group, respectively. CONCLUSIONS: The nanowire assay demonstrated excellent ability to identify HPV DNA from urine specimens. We observed an excellent agreement in the detection of high-risk HPV between paired urine and cervical samples, even with small urine sample volume.

Cytological sampling of fallopian tubes using a hysteroscopic catheter: A multi-center study.

OBJECTIVE: To assess the feasibility of a novel hysteroscopic catheter to collect fallopian tube cytologic samples and to correlate cytologic findings with histopathology. METHODS: This was a prospective, multicenter, single-arm pilot study. Women undergoing salpingo-oophorectomy for a pelvic mass suspicious for malignancy or for prevention of cancer for BRCA mutation carriers were recruited from 3 gynecologic oncology centers (October 2016-August 2017). Cytologic samples were collected from the fallopian tube using a novel FDA-cleared hysteroscopic catheter and evaluated by a pathologist blinded to surgical or pathologic findings. The correlation between cytologic results and final surgical pathology was assessed. RESULTS: Of the 50 patients enrolled, 42 were eligible. Hysteroscopies were completed in 40 patients with 78 fallopian tubes, of which 65 ostia (83%) were identified. Of these, 61 (72%) were successfully catheterized resulting in 44 (68%) cytology samples adequate for further evaluation: 5 were classified as positive (3 neoplastic and 2 malignant) and 39 as negative (34 benign and 5 reactive/atypical). A comparison of cytology results with fallopian tube histopathology showed a concordance rate of 95% (42/44). Of the two samples with discordant results, both had positive cytology but negative tubal pathology, and both were stage I ovarian cancers with malignant ovary histology. CONCLUSIONS: Deployment of the device yielded an evaluable cytologic sample in 68% of cases with a high rate of concordance with histopathology. Further evaluation of the device's ability to detect malignancy in high risk populations is warranted.

Spectrofluorimetric cytosensing of colorectal cancer cells using terbium-doped dendritic fibrous nano-silica functionalized by folic acid: A novel optical cytosensor for cancer detection.

A novel fluorescent probe for detection of HT 29 cancer cells was developed based on terbium-doped dendritic fibrous nanosilica functionalized by folic acid (Tb@KCC-1-NH2-FA). Using this probe, fluorescence signals was emitted by Tb@KCC-1-NH2-FA at 490 nm by applying 380 nm as excitation wavelength. The reported probe is based on the interaction between FA decorated on the surface of Tb@KCC-1-NH2-FA and folate receptor (FR) which is overexpressed on the surface of the most of cancer cells. Fluorescence microscopy and flow cytometry were utilized to verify the uptake of Tb@KCC-1-NH2-FA with FR-positive HT 29 cancer cells. The specificity of Tb@KCC-1-NH2-FA towards FR-positive cells was approved by staining HEK 293 cells as FR-negative cells with Tb@KCC-1-NH2-FA which obtained results approved selective differentiation of normal cells with the FA-decorated nanomaterials. The cytotoxicity of Tb@KCC-1-NH2-FA was evaluated by MTT assay which confirmed their biocompatible nature. Under optimum conditions, this cytosensor is able to detect HT 29 colon cancer from 500 to 6.5 × 103 cells/mL with lower limit of detection (LLOQ) of 500 cells/mL. Due to the room temperature stability of Tb@KCC-1-NH2-FA, this cytosensor could be developed in a simple way with exceptional specificity which may show potential applications for early stage detection of colon cancer.

Can the 72-hour rule based on "Blast/Abn Lymph" flag on Sysmex XN-10 optimize the workflow in hematology laboratory?

Despite the continuing improvement of automated blood cell counters, confirmation by blood smear examination remains the gold standard in case of anomalies. With a constant goal of standardisation, different experts committees (e.g. the French-speaking cellular hematology group (Groupe francophone d'hématologie cellulaire, GFHC and the ISLH International society for laboratory hematology) recently published criteria for microscopic analysis of blood smears. Cornet et al. evaluated the application of those criteria and propose to suppress any review for 72 hours when a "Blast/Abn lymph" flag is triggered for a sample with no abnormal cell on the microscopic review. The aims of our study were to retrospectively evaluate whether this 72-hour rule adequately operates and whether it is possible to extend the arbitrary 72-hour timeframe to 96h and 144h. To achieve this goal, 40,688 blood samples were collected from three French-speaking hospitals. 1,548 samples presented an isolated "Blast/Abn lymph" flag. Only 221 samples presented the application of the 72-hour rule at least once for our study period. We were able to extend this rule to 144 hours for 10 samples of them. All blood smears for which the rule was applied were verified and there was no abnormal cell on smears at 72 and 144 hours. In conclusion, the 72-hour rule derived from the GFHC's criteria is secure and reduces the slide review rate and thus the production costs and the turnaround time of hemogram results. Further investigations could confirm that its extension to 144 hours is also adequate.

A Clinical Trial to Verify the Efficiency of the LC-1000 Exfoliative Cell Analyzer as a New Method of Cervical Cancer Screening.

OBJECTIVE: The exfoliative cell analyzer, LC-1000 (Sysmex Corporation, Japan), is a medical device that presents the cell proliferation index and 23 research parameters as indicators of cellular proliferative potential. The objective was to evaluate the clinical usability of qualitative assessment by LC-1000 compared with cytology, the human papillomavirus (HPV) test, and histology as gold standard. STUDY DESIGN: Women that visited 3 sites between July 2015 and March 2017 were registered. The primary endpoint in this study was the comparison between LC-1000 measurement and HPV test for sensitivity and specificity for cervical intraepithelial neoplasia 2+ (CIN2+). A tree model algorithm was newly constructed by a statistical method and its relationship with histological results was evaluated. RESULTS: The sensitivity and specificity of LC-1000 were 78.3 and 74.1%, while those of the HPV test were 94.7 and 85.4%, respectively. A tree model comprising five categories was constructed. The proportion of advanced lesions was higher with the change in the rank classification results from 1 to 5. The positive predictive values of CIN2+ in the categories 4 and 5 were high. Despite the small number of subjects, cancer was undetected in categories 1 and 2. In addition, the comparison with follow-up results in 19 women assessed as CIN1 showed that the rate of progression in the categories 3-5 was 50% (7/14); progression in the categories 1 and 2 was 0% (0/5). CONCLUSIONS: LC-1000 may be useful for cervical cancer screening as an index to qualitatively evaluate CIN and cancer based on the changes in characteristics of cells.

[Clinical study of brush liquid-based cytology combined with automatic immunohistochemistry in the classification and diagnosis of lung cancer].

Objective: To explore the feasibility of bronchoscopic brushing liquid-based slide cytology combined with automatic immunocytochemistry (ICC) for pathological typing of lung cancer. Methods: A liquid-based thin-prep was prepared from 171 bronchoscopic brushing specimens of patients with pulmonary lesions. ICC was detected by automatic immunohistochemistry instrument while cytomorphological diagnosis was made. The results were compared with those of histopathological diagnosis. Results: Among 171 patients, 130 (76.0%) could be classified by cell morphology alone, including 31 squamous cell carcinomas, 44 adenocarcinomas and 55 small cell carcinomas; 162 (94.7%) could be classified by cell morphology combined with ICC, including 38 squamous cell carcinomas, 61 adenocarcinomas and 63 small cell carcinomas (P<0.001). According to the gold standard of histopathological diagnosis, the coincidence rate of cytomorphology combined with ICC was higher than that of cell morphology alone. The coincidence rate of squamous cell carcinoma was increased from 85.2% to 97.1% (P=0.093), adenocarcinoma from 92.5% to 98.0% (P<0.001), and small cell carcinoma from 96.1% to 98.3% (P=0.465). Conclusion: The combination of liquid-based thin-prep cytology and automatic immunohistochemistry can effectively improve the accuracy of pathological typing of brushing specimens under fiberoptic bronchoscopy, and provide more objective diagnostic results for clinical treatment.

Our 2018 Cancer Cytopathology Young Investigator.

BACKGROUND: The Paris System for Urine Cytopathology (the Paris System) has succeeded in making the analysis of liquid-based urine preparations more reproducible. Any algorithm seeking to automate this system must accurately estimate the nuclear-to-cytoplasmic (N:C) ratio and produce a qualitative "atypia score." The authors propose a hybrid deep-learning and morphometric model that reliably automates the Paris System. METHODS: Whole-slide images (WSI) of liquid-based urine cytology specimens were extracted from 51 negative, 60 atypical, 52 suspicious, and 54 positive cases. Morphometric algorithms were applied to decompose images to their component parts; and statistics, including the NC ratio, were tabulated using segmentation algorithms to create organized data structures, dubbed rich information matrices (RIMs). These RIM objects were enhanced using deep-learning algorithms to include qualitative measures. The augmented RIM objects were then used to reconstruct WSIs with filtering criteria and to generate pancellular statistical information. RESULTS: The described system was used to calculate the N:C ratio for all cells, generate object classifications (atypical urothelial cell, squamous cell, crystal, etc), filter the original WSI to remove unwanted objects, rearrange the WSI to an efficient, condensed-grid format, and generate pancellular statistics containing quantitative/qualitative data for every cell in a WSI. In addition to developing novel techniques for managing WSIs, a system capable of automatically tabulating the Paris System criteria also was generated. CONCLUSIONS: A hybrid deep-learning and morphometric algorithm was developed for the analysis of urine cytology specimens that could reliably automate the Paris System and provide many avenues for increasing the efficiency of digital screening for urine WSIs and other cytology preparations.

Development of a cytology-based multivariate analytical risk index for oral cancer.

OBJECTIVES: The diagnosis and management of oral cavity cancers are often complicated by the uncertainty of which patients will undergo malignant transformation, obligating close surveillance over time. However, serial biopsies are undesirable, highly invasive, and subject to inherent issues with poor inter-pathologist agreement and unpredictability as a surrogate for malignant transformation and clinical outcomes. The goal of this study was to develop and evaluate a Multivariate Analytical Risk Index for Oral Cancer (MARIO) with potential to provide non-invasive, sensitive, and quantitative risk assessments for monitoring lesion progression. MATERIALS AND METHODS: A series of predictive models were developed and validated using previously recorded single-cell data from oral cytology samples resulting in a "continuous risk score". Model development consisted of: (1) training base classification models for each diagnostic class pair, (2) pairwise coupling to obtain diagnostic class probabilities, and (3) a weighted aggregation resulting in a continuous MARIO. RESULTS AND CONCLUSIONS: Diagnostic accuracy based on optimized cut-points for the test dataset ranged from 76.0% for Benign, to 82.4% for Dysplastic, 89.6% for Malignant, and 97.6% for Normal controls for an overall MARIO accuracy of 72.8%. Furthermore, a strong positive relationship with diagnostic severity was demonstrated (Pearson's coefficient = 0.805 for test dataset) as well as the ability of the MARIO to respond to subtle changes in cell composition. The development of a continuous MARIO for PMOL is presented, resulting in a sensitive, accurate, and non-invasive method with potential for enabling monitoring disease progression, recurrence, and the need for therapeutic intervention of these lesions.

Comparison of nylon-flocked swab and Dacron swab cytology for anal HSIL detection in transgender women and gay, bisexual, and other men who have sex with men.

BACKGROUND: An anal histological high-grade squamous intraepithelial lesion (hHSIL) is an anal cancer precursor. Experts recommend Dacron swab anal cytology as a primary screen for anal hHSILs, especially among human immunodeficiency virus-infected and -uninfected men who have sex with men (MSM). Studies have shown that Dacron cytology inaccurately predicts anal hHSILs and results in unnecessary diagnostic procedures. Nylon-flocked (NF) swabs have been shown to trap pathogens and cells well. Thus, this study compared test characteristics of anal cytology using NF and Dacron swab collection protocols to predict anal hHSILs. METHODS: A single-visit, randomized clinical trial compared NF and Dacron swab anal cytology specimens to predict high-resolution anoscopy and biopsy-diagnosed anal hHSILs. Data for 326 gay men, bisexual men, other MSM, and male-to-female transgender women contributed descriptive and tabular statistics with which unadjusted and fully adjusted logistic regression models were constructed. The models estimated the odds of hHSILs, test accuracy (area under the curve [AUC]) and sensitivity, and specificity as well as the positive and negative predictive values of abnormal NF and Dacron cytology for predicting hHSILs. RESULTS: In the fully adjusted model, the sensitivities for NF and Dacron cytology were nearly equal (48% vs 47%), but the specificity was higher with NF cytology (76% vs 69%). Comparisons of the areas under receiver operating characteristic curves showed that NF cytology alone predicted hHSILs better than the covariate model (AUC, 0.69 vs 0.63; P = .02), but NF and Dacron cytology comparisons showed no statistically significant differences (AUC, 0.69 vs 0.67; P = .3). CONCLUSIONS: NF cytology and Dacron cytology provide modest sensitivity, but NF cytology has higher specificity and accuracy, and this is important for lowering the costs of population-based screening.

Automating the Paris System for urine cytopathology-A hybrid deep-learning and morphometric approach.

BACKGROUND: The Paris System for Urine Cytopathology (the Paris System) has succeeded in making the analysis of liquid-based urine preparations more reproducible. Any algorithm seeking to automate this system must accurately estimate the nuclear-to-cytoplasmic (N:C) ratio and produce a qualitative "atypia score." The authors propose a hybrid deep-learning and morphometric model that reliably automates the Paris System. METHODS: Whole-slide images (WSI) of liquid-based urine cytology specimens were extracted from 51 negative, 60 atypical, 52 suspicious, and 54 positive cases. Morphometric algorithms were applied to decompose images to their component parts; and statistics, including the NC ratio, were tabulated using segmentation algorithms to create organized data structures, dubbed rich information matrices (RIMs). These RIM objects were enhanced using deep-learning algorithms to include qualitative measures. The augmented RIM objects were then used to reconstruct WSIs with filtering criteria and to generate pancellular statistical information. RESULTS: The described system was used to calculate the N:C ratio for all cells, generate object classifications (atypical urothelial cell, squamous cell, crystal, etc), filter the original WSI to remove unwanted objects, rearrange the WSI to an efficient, condensed-grid format, and generate pancellular statistics containing quantitative/qualitative data for every cell in a WSI. In addition to developing novel techniques for managing WSIs, a system capable of automatically tabulating the Paris System criteria also was generated. CONCLUSIONS: A hybrid deep-learning and morphometric algorithm was developed for the analysis of urine cytology specimens that could reliably automate the Paris System and provide many avenues for increasing the efficiency of digital screening for urine WSIs and other cytology preparations.

Telecytology implementation: Deployment of telecytology for rapid on-site evaluations at an Academic Medical Center.

INTRODUCTION: There are limited publications that address technical and practical informatics considerations when implementing telecytology for rapid on-site evaluation (ROSE). Our aim was to share the experience of deploying telecytology for ROSE at our institution. MATERIALS AND METHODS: Key informatics issues relevant to adopting telecytology for ROSE at our institution were appraised including workflow, information technology (IT), validation, training, and quality assurance (QA). RESULTS: A dynamic telemicroscopy solution was selected that required trained cytotechnologists to attend on-site procedures for ROSE. For validation 60 cases were reviewed using the first camera at each facility, but only 20 cases to validate subsequent cameras. A concordance rate of >90% between ROSE interpretation performed digitally to original interpretations was required for clinical validation. After reviewing 440 cases from two comparable time periods before and after implementation, employing telecytology was shown to decrease cytopathologists' work time per ROSE case from an average of 20.95 min per case to 2.91 min per case (86% time savings). The non-diagnostic rate for traditional ROSE was 7.7% compared with 4.1% after the implementation of telecytology, and the deferral rate went from 43.6% for traditional ROSE to 44.1% with telecytology. Traditional ROSE diagnoses correlated with final diagnoses in 91.8% cases, compared to 95.5% with telecytology. CONCLUSIONS: Challenges when implementing telecytology for ROSE included technical issues, workflow concerns, and incorporating trainees into daily practice. The end result of our implementation was the adoption of an innovative way to deliver a ROSE service that maximised efficiency for cytopathologists without compromising diagnostic performance.