Staining, magnification, and algorithmic conditions for highly accurate cell detection and cell classification by deep learning.

OBJECTIVES: Research into cytodiagnosis has seen an active exploration of cell detection and classification using deep learning models. We aimed to clarify the challenges of magnification, staining methods, and false positives in creating general purpose deep learning-based cytology models. METHODS: Using 11 types of human cancer cell lines, we prepared Papanicolaou- and May-Grünwald-Giemsa (MGG)-stained specimens. We created deep learning models with different cell types, staining, and magnifications from each cell image using the You Only Look Once, version 8 (YOLOv8) algorithm. Detection and classification rates were calculated to compare the models. RESULTS: The classification rates of all the created models were over 95.9%. The highest detection rates of the Papanicolaou and MGG models were 92.3% and 91.3%, respectively. The highest detection rates of the object detection and instance segmentation models, which were 11 cell types with Papanicolaou staining, were 94.6% and 91.7%, respectively. CONCLUSIONS: We believe that the artificial intelligence technology of YOLOv8 has sufficient performance for applications in screening and cell classification in clinical settings. Conducting research to demonstrate the efficacy of YOLOv8 artificial intelligence technology on clinical specimens is crucial for overcoming the unique challenges associated with cytology.

Effect of liquid-based cytology fixing solution on immunocytochemistry: Efficacy of antigen retrieval in cytologic specimens.

BACKGROUND: Immunocytochemistry (ICC) is an indispensable technique to improve diagnostic accuracy. ICC using liquid-based cytology (LBC)-fixed specimens has been reported. However, problems may arise if the samples are not fixed appropriately. We investigated the relationship between the LBC fixing solution and ICC and the usefulness of antigen retrieval (AR) in LBC specimens. METHODS: Specimens were prepared from five types of LBC-fixed samples using cell lines and the SurePath™ method. ICC was performed using 13 antibodies and analyzed by counting the number of positive cells in the immunocytochemically stained specimens. RESULTS: Insufficient reactivity was observed using ICC without heat-induced AR (HIAR) in nuclear antigens. The number of positive cells increased in ICC with HIAR. The percentage of positive cells was lower in CytoRich™ Blue samples for Ki-67 and in CytoRich™ Red and TACAS™ Ruby samples for estrogen receptor and p63 than in the other samples. For cytoplasmic antigens, the percentage of positive cells for no-HIAR treatment specimens was low in the three antibodies used. In cytokeratin 5/6, the number of positive cells increased in all LBC specimens with HIAR, and the percentage of positive cells in CytoRich™ Red and TACAS™ Ruby samples was significantly lower (p < .01). For cell membrane antigens, CytoRich™ Blue samples had a lower percentage of positive cells than the other LBC-fixed samples. CONCLUSION: The combination of detected antigen, used cells, and fixing solution may have different effects on immunoreactivity. ICC using LBC specimens is a useful technique, but the staining conditions should be examined before performing ICC.

Relationship between a deep learning model and liquid-based cytological processing techniques.

OBJECTIVE: Artificial intelligence (AI)-based cytopathology studies conducted using deep learning have enabled cell detection and classification. Liquid-based cytology (LBC) has facilitated the standardisation of specimen preparation; however, cytomorphology varies according to the LBC processing technique used. In this study, we elucidated the relationship between two LBC techniques and cell detection and classification using a deep learning model. METHODS: Cytological specimens were prepared using the ThinPrep and SurePath methods. The accuracy of cell detection and cell classification was examined using the one- and five-cell models, which were trained with one and five cell types, respectively. RESULTS: When the same LBC processing techniques were used for the training and detection preparations, the cell detection and classification rates were high. The model trained on ThinPrep preparations was more accurate than that trained on SurePath. When the preparation types used for training and detection were different, the accuracy of cell detection and classification was significantly reduced (P < 0.01). The model trained on both ThinPrep and SurePath preparations exhibited slightly reduced cell detection and classification rates but was highly accurate. CONCLUSIONS: For the two LBC processing techniques, cytomorphology varied according to cell type; this difference affects the accuracy of cell detection and classification by deep learning. Therefore, for highly accurate cell detection and classification using AI, the same processing technique must be used for both training and detection. Our assessment also suggests that a deep learning model should be constructed using specimens prepared via a variety of processing techniques to construct a globally applicable AI model.

Effect of Specimen Processing Technique on Cell Detection and Classification by Artificial Intelligence.

OBJECTIVES: Cytomorphology is known to differ depending on the processing technique, and these differences pose a problem for automated diagnosis using deep learning. We examined the as-yet unclarified relationship between cell detection or classification using artificial intelligence (AI) and the AutoSmear (Sakura Finetek Japan) and liquid-based cytology (LBC) processing techniques. METHODS: The "You Only Look Once" (YOLO), version 5x, algorithm was trained on the AutoSmear and LBC preparations of 4 cell lines: lung cancer (LC), cervical cancer (CC), malignant pleural mesothelioma (MM), and esophageal cancer (EC). Detection and classification rates were used to evaluate the accuracy of cell detection. RESULTS: When preparations of the same processing technique were used for training and detection in the 1-cell (1C) model, the AutoSmear model had a higher detection rate than the LBC model. When different processing techniques were used for training and detection, detection rates of LC and CC were significantly lower in the 4-cell (4C) model than in the 1C model, and those of MM and EC were approximately 10% lower in the 4C model. CONCLUSIONS: In AI-based cell detection and classification, attention should be paid to cells whose morphologies change significantly depending on the processing technique, further suggesting the creation of a training model.

Immunocytochemical Reactivity Comparison between Formalin-Fixed and Liquid-Based Cytology-Fixed Specimens.

INTRODUCTION: Liquid-based cytology (LBC)-fixed samples can be used for preparing multiple specimens of the same quality and for immunocytochemistry (ICC); however, LBC fixing solutions affect immunoreactivity. Therefore, in this study, we examined the effect of LBC fixing solutions on immunoreactivity. METHODS: Samples were cell lines, and specimens were prepared from cell blocks of 10% neutral buffered formalin (NBF)-fixed samples and the four types of LBC-fixed samples: PreservCyt®, CytoRich™ Red, CytoRich™ Blue, and TACAS™ Ruby, which were post-fixed with NBF. ICC was performed using 24 different antibodies, and immunocytochemically stained specimens were analyzed for the percentage of positive cells. RESULTS: Immunoreactivity differed according to the type of antigen detected. For nuclear antigens, the highest percentage of positive cells of Ki-67, WT-1, ER, and p63 was observed in the NBF-fixed samples, and the highest percentage of positive cells of p53, TTF-1, and PgR was observed in the TACAS™ Ruby samples. For cytoplasmic antigens, the percentage of positive cells of CK5/6, Vimentin, and IMP3 in LBC-fixed samples was higher than or similar to that in NBF-fixed samples. The percentage of positive cells of CEA was significantly lower in CytoRich™ Red and CytoRich™ Blue samples than in the NBF-fixed sample (p < 0.01). Among the cell membrane antigens, the percentage of positive cells of Ber-EP4, CD10, and D2-40 was the highest in NBF-fixed samples and significantly lower in CytoRich™ Red and CytoRich™ Blue samples than that in NBF-fixed samples (p < 0.01). The NBF-fixed and LBC-fixed samples showed no significant differences in the percentage of positive cells of CA125 and EMA. DISCUSSION/CONCLUSION: ICC using LBC-fixed samples showed the same immunoreactivity as NBF-fixed samples when performed on cell block specimens post-fixed with NBF. The percentage of positive cells increased or decreased based on the type of fixing solution depending on the amount of antigen in the cells. Further, the detection rate of ICC with LBC-fixed samples varied according to the type of antibody and the amount of antigen in the cells. Therefore, we propose that ICC using LBC-fixed samples, including detection methods, should be carefully performed.

Relationship between Liquid-Based Cytology Preservative Solutions and Artificial Intelligence: Liquid-Based Cytology Specimen Cell Detection Using YOLOv5 Deep Convolutional Neural Network.

INTRODUCTION: Deep learning is a subset of machine learning that has contributed to significant changes in feature extraction and image classification and is being actively researched and developed in the field of cytopathology. Liquid-based cytology (LBC) enables standardized cytological preparation and is also applied to artificial intelligence (AI) research, but cytological features differ depending on the LBC preservative solution types. In this study, the relationship between cell detection by AI and the type of preservative solution used was examined. METHODS: The specimens were prepared from five preservative solutions of LBC and stained using the Papanicolaou method. The YOLOv5 deep convolutional neural network algorithm was used to create a deep learning model for each specimen, and a BRCPT model from five specimens was also created. Each model was compared to the specimen types used for detection. RESULTS: Among the six models, a difference in the detection rate of approximately 25% was observed depending on the detected specimen, and within specimens, a difference in the detection rate of approximately 20% was observed depending on the model. The BRCPT model had little variation in the detection rate depending on the type of the detected specimen. CONCLUSIONS: The same cells were treated with different preservative solutions, the cytologic features were different, and AI clarified the difference in cytologic features depending on the type of solution. The type of preservative solution used for training and detection had an extreme influence on cell detection using AI. Although the accuracy of the deep learning model is important, it is necessary to understand that cell morphology differs depending on the type of preservative solution, which is a factor affecting the detection rate of AI.

Health Effects of Drinking Water Produced from Deep Sea Water: A Randomized Double-Blind Controlled Trial.

Global trends focus on a balanced intake of foods and beverages to maintain health. Drinking water (MIU; hardness = 88) produced from deep sea water (DSW) collected offshore of Muroto, Japan, is considered healthy. We previously reported that the DSW-based drinking water (RDSW; hardness = 1000) improved human gut health. The aim of this randomized double-blind controlled trial was to assess the effects of MIU on human health. Volunteers were assigned to MIU (n = 41) or mineral water (control) groups (n = 41). Participants consumed 1 L of either water type daily for 12 weeks. A self-administered questionnaire was administered, and stool and urine samples were collected throughout the intervention. We measured the fecal biomarkers of nine short-chain fatty acids (SCFAs) and secretory immunoglobulin A (sIgA), as well as urinary isoflavones. In the MIU group, concentrations of three major SCFAs and sIgA increased postintervention. MIU intake significantly affected one SCFA (butyric acid). The metabolic efficiency of daidzein-to-equol conversion was significantly higher in the MIU group than in the control group throughout the intervention. MIU intake reflected the intestinal environment through increased production of three major SCFAs and sIgA, and accelerated daidzein-to-equol metabolic conversion, suggesting the beneficial health effects of MIU.

Characterizing the Effect of Processing Technique and Solution Type on Cytomorphology Using Liquid-Based Cytology.

INTRODUCTION: Liquid-based cytology (LBC) is increasingly used for nongynecologic applications. However, the cytological preparation of LBC specimens is influenced by the processing technique and the preservative used. In this study, the influence of the processing techniques and preservatives on cell morphology was examined mathematically and statistically. METHODS: Cytological specimens were prepared using the ThinPrep (TP), SurePath (SP), and AutoSmear methods, with 5 different preservative solutions. The cytoplasmic and nuclear areas of Papanicolaou-stained specimens were measured for all samples. RESULTS: The cytoplasmic and nuclear areas were smaller in cells prepared using the 2 LBC methods, compared to that prepared using the AutoSmear method, irrespective of the preservative used. The cytoplasmic and nuclear areas of cells prepared using the SP method were smaller than those of cells prepared using the TP method, irrespective of the preservative used. There were fewer differences among the cytoplasmic areas of cells prepared with different preservative solutions using the TP method; however, the cytoplasmic areas of cells prepared using the SP method changed with the preservative solution used. CONCLUSIONS: The most significant difference affecting the cytoplasmic and nuclear areas was the processing technique. The TP method increased the cytoplasmic and nuclear areas, while the methanol-based PreservCyt solution enabled the highest enlargement of the cell. LBC is a superior preparation technique for standardization of the specimens. Our results offer a better understanding of methods suitable for specimen preparation for developing precision AI-based diagnosis in cytology.

Drinking Refined Deep-Sea Water Improves the Gut Ecosystem with Beneficial Effects on Intestinal Health in Humans: A Randomized Double-Blind Controlled Trial.

World health trends are focusing on a balanced food and beverage intake for healthy life. Refined deep-sea water (RDSW), obtained from deep-sea water collected offshore in Muroto (Japan), is mineral-rich drinking water. We previously reported that drinking RDSW improves human gut health. Here, we analyzed the effect of drinking RDSW on the gut ecosystem to understand this effect. This was a randomized double-blind controlled trial. Ninety-eight healthy adults were divided into two groups: RDSW or mineral water (control). The participants consumed 1 L of either water type daily for 12 weeks. A self-administered questionnaire and stool and urine samples were collected through the intervention. The following were determined: fecal biomarkers of secretory immunoglobulin A (sIgA), five putrefactive products, and nine short-chain-fatty-acids (SCFAs) as the primary outcomes; and three urinary isoflavones and the questionnaire as secondary outcomes. In post-intervention in the RDSW group, we found increased concentrations of five SCFAs and decreased concentrations of phenol and sIgA (p < 0.05). The multiple logistic analysis demonstrated that RDSW significantly affected two biomarkers (acetic and 3-methylbutanoic acids) of the five SCFAs mentioned above (p < 0.05). Similarly, the concentrations of urinary isoflavones tended to increase in post-intervention in the RDSW group. Constipation was significantly alleviated in the RDSW group (94%) compared with the control group (60%). Drinking RDSW improves the intestinal environment, increasing fecal SCFAs and urinary isoflavones, which leads to broad beneficial effects in human.

Characterizing the Effect of Automated Cell Sorting Solutions on Cytomorphological Changes.

INTRODUCTION: Liquid-based cytology has become a widely adopted, automated screening system for gynecologic and nongynecologic cytology. Automated screening systems function by distinguishing atypical cells based on their cytoplasmic and nuclear areas, densitometric measurement, and so on. However, the morphological influence of the washing solution has not been fully considered. Here, we examined the morphological effect and temporal change resulting from saving the cytologic samples in various solutions. METHODS: Cytologic specimens were obtained from the ascites (AS) of patients with peritoneal cancer. Various solutions of a physiological saline, a Ringer's solution, a low-molecular dextran L injection, VOLUVEN 6% solution, MIXID L injection (ML), RPMI-1640 medium, and horse serum (HS) were added to aliquot sediments. All samples were refrigerated at 4°C, and aliquots were subsequently processed at specific time points (0, 1, 2, 4, 7, and 14 days). For all samples, cytoplasmic and nuclear size of the Papanicolaou-stained specimens were measured. RESULTS: In terms of cytoplasmic and nuclear areas, samples stored in ML and HS showed no significant difference compared to the AS sample; in contrast, the other samples were significantly larger in both cytoplasmic and nuclear areas than the AS sample. In examining the temporal change among the solutions, we found that the cytoplasms and nuclei became small over the time course for all of the tested solutions. CONCLUSION: We showed that cells swell in the solution after 1 h of storage and contract as time progresses. Together, our findings have important implications for how mathematical analysis is applied during the automated screening process.

Effects of the Storage Solution Type and Prolonged Storage on the Immunoreactivity of Cells.

INTRODUCTION: In effusion cytology, immunocytochemistry is a useful staining approach to provide important information for diagnosis. Effusion cytology is performed not only for pleural effusions and ascites but also for peritoneal and needle washing from fine needle aspirations or instruments. Although various solutions are used for washing cytology, the effect of the solution type on immunocytochemical reactivity is not fully understood. In this study, we examined the immunocytochemical reactivity of cytological samples after storage in various solutions. METHODS: Cell block specimens were obtained from ascites of patients with peritoneal cancer and pleural effusions of patients with diffuse malignant mesothelioma. Various solutions, including physiological saline (PS), Ringer solution, a low-molecular-weight dextran L injection, Voluven 6% solution, Mixid L injection, RPMI-1640 medium, and horse serum were added to the sediment layers of aliquots. All samples were kept at 4°C, and aliquots were subsequently processed at specific time points (0, 1, 2, 4, 7, and 14 days). Formalin-fixed, paraffin-embedded, cell block samples were prepared for immunocytochemical staining. Immunocytochemical results were analyzed for differences in the percentages of positive cells, using the effusion sample stored for 1 h as standard (100%). RESULTS: For all solutions other than PS, the median and central 50% of values were <100% (with respect to the effusion sample as a standard) after 1 h of storage. Immunoreactivity decreased for most solutions as time progressed. CONCLUSION: Of note, immunocytochemistry results obtained using a washing solution are different from those using an effusion sample. For cytology, when a washing solution was used or when a sample was stored for a long time, the accuracy of the immunocytochemical results was low.

[Multivariate Analyses on Clinical Utility and Sources of Variation of Serum Presepsin as a Diagnostic Marker for Sepsis].

OBJECTIVE: Clinical utility of a new marker for sepsis, presepsin, was evaluated by use of a case-control study design. METHOD: Enrolled in the study were seventy-one consecutive cases for whom blood culture was ordered in suspicion of sepsis. After the culture, 36 subjects were diagnosed as having a state of sepsis (S group) and 35 were denied of sepsis (NS group). The serum level of presepsin was measured together with basic chemistry tests and complete blood counts at the time of diagnosis. RESULTS: Median serum presepsin for the two groups were 1,602 and 586 pg/mL, respectively. The difference was significant by Mann-Whitney test (P < 0.001). Logistic regression analysis was performed to evaluate contribution of presepsin in diagnosing sepsis in comparison with other markers for septic state. The result showed that presepsin was most powerful in predicting sepsis together with monocyte count percent (Mo). The diagnostic accuracy by use of logistic equation including both presepsin and Mo was 0.86 in terms of area under ROC curve (AUC), whereas AUC by use of an equation with presepsin alone was 0.80. Additionally, multiple regression analysis was performed to evaluate sources of variation of presepsin. It revealed that serum albumin and eGFR were negatively associated with serum level of presepsin. CONCLUSION: It is recommended to look at Mo together with presepsin in the diagnosis of sepsis. Serum level of presepsin is raised in the presence of renal dysfunction and/or hypoalbuminemia.

A new self-partition clustering method for robust identification of subsets with heterogeneous size and density and its clinical application to leukocyte differential counting.

BACKGROUND: Identification of clusters in 2-dimensional scatterplots generated by hematology analyzer is a classical challenge. Conventional clustering algorithms fail to process cases with complicated mixtures of overlapping clusters and noise. METHOD: A new method was developed that features an image processing algorithm for rational identification of initial clusters and a self-partition clustering (SPC) algorithm with iterative truncation-correction (ITC) method to handle overlapping and noise. All clusters are assumed to follow bivariate Gaussian distributions with specified means, SDs, and correlation coefficient. While, each data point is assumed to belong to all clusters but with different proportions according to the likelihood of belonging to each cluster (computed by the Mahalanobis distance) and the data size of the cluster. Bivariate cluster statistics are computed in consideration of a weight factor determined cluster by cluster by each data point. In the computation, the ITC method minimizes the effect of overlapping and data. RESULTS: Performance of SPC/ITC method was evaluated by its application to differential leukocyte counting. It showed comparable performance with manual counting and much better performance than the commonly used expectation maximum algorithm. CONCLUSION: The SPC/ITC method showed superior performance in situations with overlapping and low-density clusters such as leukopenia or leukocytosis.