A study of criteria for grading follicular lymphoma using a cell type classifier from pathology images based on complementary-label learning.

We propose a criterion for grading follicular lymphoma that is consistent with the intuitive evaluation, which is conducted by experienced pathologists. A criterion for grading follicular lymphoma is defined by the World Health Organization (WHO) based on the number of centroblasts and centrocytes within the field of view. However, the WHO criterion is not often used in clinical practice because it is impractical for pathologists to visually identify the cell type of each cell and count the number of centroblasts and centrocytes. Hence, based on the widespread use of digital pathology, we make it practical to identify and count the cell type by using image processing and then construct a criterion for grading based on the number of cells. Here, the problem is that labeling the cell type is not easy even for experienced pathologists. To alleviate this problem, we build a new dataset for cell type classification, which contains the pathologists' confusion records during labeling, and we construct the cell type classifier using complementary-label learning from this dataset. Then we propose a criterion based on the composition ratio of cell types that is consistent with the pathologists' grading. Our experiments demonstrate that the classifier can accurately identify cell types and the proposed criterion is more consistent with the pathologists' grading than the current WHO criterion.

Discovery of Novel Bicyclic Pyrazoles as Potent PIP5K1C Inhibitors.

Phosphatidylinositol-4,5-bisphosphate (PI(4,5)P2) is generated by phosphatidylinositol 4-phosphate 5-kinases (PIP5Ks) from phosphatidylinositol 4-phosphate (PI4P). Structurally diverse and selective inhibitors against PIP5Ks are required to further elucidate the therapeutic potential for PIP5K inhibition, although the effects of PIP5K inhibition on various diseases and their symptoms, such as cancer and chronic pain, have been reported. Our medicinal chemistry efforts led to novel and potent PIP5K1C inhibitors. Compounds 30 and 33 not only showed potent activity but also demonstrated low total clearance in mice and high levels of kinase selectivity. These compounds might serve as tools to further elucidate the complex biology and therapeutic potential of PIP5K inhibition.

Multimodal Gated Mixture of Experts Using Whole Slide Image and Flow Cytometry for Multiple Instance Learning Classification of Lymphoma.

In this study, we present a deep-learning-based multimodal classification method for lymphoma diagnosis in digital pathology, which utilizes a whole slide image (WSI) as the primary image data and flow cytometry (FCM) data as auxiliary information. In pathological diagnosis of malignant lymphoma, FCM serves as valuable auxiliary information during the diagnosis process, offering useful insights into predicting the major class (superclass) of subtypes. By incorporating both images and FCM data into the classification process, we can develop a method that mimics the diagnostic process of pathologists, enhancing the explainability. In order to incorporate the hierarchical structure between superclasses and their subclasses, the proposed method utilizes a network structure that effectively combines the mixture of experts (MoE) and multiple instance learning (MIL) techniques, where MIL is widely recognized for its effectiveness in handling WSIs in digital pathology. The MoE network in the proposed method consists of a gating network for superclass classification and multiple expert networks for (sub)class classification, specialized for each superclass. To evaluate the effectiveness of our method, we conducted experiments involving a six-class classification task using 600 lymphoma cases. The proposed method achieved a classification accuracy of 72.3%, surpassing the 69.5% obtained through the straightforward combination of FCM and images, as well as the 70.2% achieved by the method using only images. Moreover, the combination of multiple weights in the MoE and MIL allows for the visualization of specific cellular and tumor regions, resulting in a highly explanatory model that cannot be attained with conventional methods. It is anticipated that by targeting a larger number of classes and increasing the number of expert networks, the proposed method could be effectively applied to the real problem of lymphoma diagnosis.

The Detection of Neoplastic Cells Using Objective Cytomorphologic Parameters in Malignant Lymphoma.

Pathologic evaluation is the most crucial method for diagnosing malignant lymphomas. However, there are no established diagnostic criteria for evaluating pathologic morphology. We manually circled cell nuclei in the lesions of 10 patients with diffuse large B-cell lymphoma (DLBCL), follicular lymphoma, and reactive lymphadenitis. Seventeen parameters related to nuclear shape, color, and other characteristics were measured. We attempted to compare the statistical differences between these subtypes and extract distinctive disease-specific populations on the basis of these parameters. Statistically significant differences were observed between the different types of lymphoma for many of the 17 parameters. Through t-distributed stochastic neighbor embedding analysis, we extracted a cluster of cells that showed distinctive features of DLBCL and were not found in follicular lymphoma or reactive lymphadenitis. We created a decision tree to identify the characteristics of the cells within that cluster. Based on a 5-fold cross-validation study, the average sensitivity, specificity, and accuracy obtained were 84.1%, 98.4%, and 97.3%, respectively. A similar result was achieved using a validation experiment. Important parameters that indicate the features of DLBCL include Area, ConcaveCount, MaxGray, and ModeGray. By quantifying pathologic morphology, it was possible to objectively represent the cell morphology specific to each lymphoma subtype using quantitative indicators. The quantified morphologic information has the potential to serve as a reproducible and flexible diagnostic tool.

Deep learning based identification of pituitary adenoma on surgical endoscopic images: a pilot study.

Accurate tumor identification during surgical excision is necessary for neurosurgeons to determine the extent of resection without damaging the surrounding tissues. No conventional technologies have achieved reliable performance for pituitary adenomas. This study proposes a deep learning approach using intraoperative endoscopic images to discriminate pituitary adenomas from non-tumorous tissue inside the sella turcica. Static images were extracted from 50 intraoperative videos of patients with pituitary adenomas. All patients underwent endoscopic transsphenoidal surgery with a 4 K ultrahigh-definition endoscope. The tumor and non-tumorous tissue within the sella turcica were delineated on static images. Using intraoperative images, we developed and validated deep learning models to identify tumorous tissue. Model performance was evaluated using a fivefold per-patient methodology. As a proof-of-concept, the model's predictions were pathologically cross-referenced with a medical professional's diagnosis using the intraoperative images of a prospectively enrolled patient. In total, 605 static images were obtained. Among the cropped 117,223 patches, 58,088 were labeled as tumors, while the remaining 59,135 were labeled as non-tumorous tissues. The evaluation of the image dataset revealed that the wide-ResNet model had the highest accuracy of 0.768, with an F1 score of 0.766. A preliminary evaluation on one patient indicated alignment between the ground truth set by neurosurgeons, the model's predictions, and histopathological findings. Our deep learning algorithm has a positive tumor discrimination performance in intraoperative 4-K endoscopic images in patients with pituitary adenomas.

Generalized Low-Rank Update: Model Parameter Bounds for Low-Rank Training Data Modifications.

In this study, we have developed an incremental machine learning (ML) method that efficiently obtains the optimal model when a small number of instances or features are added or removed. This problem holds practical importance in model selection, such as cross-validation (CV) and feature selection. Among the class of ML methods known as linear estimators, there exists an efficient model update framework, the low-rank update, that can effectively handle changes in a small number of rows and columns within the data matrix. However, for ML methods beyond linear estimators, there is currently no comprehensive framework available to obtain knowledge about the updated solution within a specific computational complexity. In light of this, our study introduces a the generalized low-rank update (GLRU) method, which extends the low-rank update framework of linear estimators to ML methods formulated as a certain class of regularized empirical risk minimization, including commonly used methods such as support vector machines and logistic regression. The proposed GLRU method not only expands the range of its applicability but also provides information about the updated solutions with a computational complexity proportional to the number of data set changes. To demonstrate the effectiveness of the GLRU method, we conduct experiments showcasing its efficiency in performing cross-validation and feature selection compared to other baseline methods.

Discovery of Novel Substrate-Competitive Lysine Methyltransferase G9a Inhibitors as Anticancer Agents.

Identification of structurally novel inhibitors of lysine methyltransferase G9a has been a subject of intense research in cancer epigenetics. Starting with the high-throughput screening (HTS) hit rac-10a obtained from the chemical library of the University of Tokyo Drug Discovery Initiative, the structure-activity relationship of the unique substrate-competitive inhibitors was established with the help of X-ray crystallography and fragment molecular orbital (FMO) calculations for the ligand-protein interaction. Further optimization of the in vitro characteristics and drug metabolism and pharmacokinetics (DMPK) properties led to the identification of 26j (RK-701), which is a structurally distinct potent inhibitor of G9a/GLP (IC50 = 27/53 nM). Compound 26j exhibited remarkable selectivity against other related methyltransferases, dose-dependent attenuation of cellular H3K9me2 levels, and tumor growth inhibition in MOLT-4 cells in vitro. Moreover, compound 26j showed inhibition of tumor initiation and growth in a carcinogen-induced hepatocellular carcinoma (HCC) in vivo mouse model without overt acute toxicity.

A specific G9a inhibitor unveils BGLT3 lncRNA as a universal mediator of chemically induced fetal globin gene expression.

Sickle cell disease (SCD) is a heritable disorder caused by ß-globin gene mutations. Induction of fetal γ-globin is an established therapeutic strategy. Recently, epigenetic modulators, including G9a inhibitors, have been proposed as therapeutic agents. However, the molecular mechanisms whereby these small molecules reactivate γ-globin remain unclear. Here we report the development of a highly selective and non-genotoxic G9a inhibitor, RK-701. RK-701 treatment induces fetal globin expression both in human erythroid cells and in mice. Using RK-701, we find that BGLT3 long non-coding RNA plays an essential role in γ-globin induction. RK-701 selectively upregulates BGLT3 by inhibiting the recruitment of two major γ-globin repressors in complex with G9a onto the BGLT3 gene locus through CHD4, a component of the NuRD complex. Remarkably, BGLT3 is indispensable for γ-globin induction by not only RK-701 but also hydroxyurea and other inducers. The universal role of BGLT3 in γ-globin induction suggests its importance in SCD treatment.

Transformer-based personalized attention mechanism for medical images with clinical records.

In medical image diagnosis, identifying the attention region, i.e., the region of interest for which the diagnosis is made, is an important task. Various methods have been developed to automatically identify target regions from given medical images. However, in actual medical practice, the diagnosis is made based on both the images and various clinical records. Consequently, pathologists examine medical images with prior knowledge of the patients and the attention regions may change depending on the clinical records. In this study, we propose a method, called the Personalized Attention Mechanism (PersAM) method, by which the attention regions in medical images according to the clinical records. The primary idea underlying the PersAM method is the encoding of the relationships between medical images and clinical records using a variant of the Transformer architecture. To demonstrate the effectiveness of the PersAM method, we applied it to a large-scale digital pathology problem involving identifying the subtypes of 842 malignant lymphoma patients based on their gigapixel whole-slide images and clinical records.

Case-based similar image retrieval for weakly annotated large histopathological images of malignant lymphoma using deep metric learning.

In the present study, we propose a novel case-based similar image retrieval (SIR) method for hematoxylin and eosin (H&E) stained histopathological images of malignant lymphoma. When a whole slide image (WSI) is used as an input query, it is desirable to be able to retrieve similar cases by focusing on image patches in pathologically important regions such as tumor cells. To address this problem, we employ attention-based multiple instance learning, which enables us to focus on tumor-specific regions when the similarity between cases is computed. Moreover, we employ contrastive distance metric learning to incorporate immunohistochemical (IHC) staining patterns as useful supervised information for defining appropriate similarity between heterogeneous malignant lymphoma cases. In the experiment with 249 malignant lymphoma patients, we confirmed that the proposed method exhibited higher evaluation measures than the baseline case-based SIR methods. Furthermore, the subjective evaluation by pathologists revealed that our similarity measure using IHC staining patterns is appropriate for representing the similarity of H&E stained tissue images for malignant lymphoma.

Subtype classification of malignant lymphoma using immunohistochemical staining pattern.

PURPOSE: For the image classification problem, the construction of appropriate training data is important for improving the generalization ability of the classifier in particular when the size of the training data is small. We propose a method that quantitatively evaluates the typicality of a hematoxylin-and-eosin (H&E)-stained tissue slide from a set of immunohistochemical (IHC) stains and applies the typicality to instance selection for the construction of classifiers that predict the subtype of malignant lymphoma to improve the generalization ability. METHODS: We define the typicality of the H&E-stained tissue slides by the ratio of the probability density of the IHC staining patterns on low-dimensional embedded space. Employing a multiple-instance-learning-based convolutional neural network for the construction of the subtype classifier without the annotations indicating cancerous regions in whole slide images, we select the training data by referring to the evaluated typicality to improve the generalization ability. We demonstrate the effectiveness of the instance selection based on the proposed typicality in a three-class subtype classification of 262 malignant lymphoma cases. RESULTS: In the experiment, we confirmed that the subtypes of typical instances could be predicted more accurately than those of atypical instances. Furthermore, it was confirmed that instance selection for the training data based on the proposed typicality improved the generalization ability of the classifier, wherein the classification accuracy was improved from 0.664 to 0.683 compared with the baseline method when the training data was constructed focusing on typical instances. CONCLUSION: The experimental results showed that the typicality of the H&E-stained tissue slides computed from IHC staining patterns is useful as a criterion for instance selection to enhance the generalization ability, and this typicality could be employed for instance selection under some practical limitations.

Mixed HCV Infection of Genotype 1B and Other Genotypes Influences Non-response during Daclatasvir + Asunaprevir Combination Therapy.

Daclatasvir (DCV) + asunaprevir (ASV) combination therapy has become available for patients with hepatitis C virus (HCV) serogroup 1 infection. We studied the efficacy of this therapy by focusing on the factors associated with sustained virological responses (SVR) including resistance-associated variants (RAVs) and mixed infection of different HCV genotypes. We enrolled 951 HCV serogroup 1-positive patients who received this combination therapy at our hospital or affiliated hospitals. The presence of RAVs in non-structural (NS) regions 3 and 5A was analyzed by direct sequencing. HCV genotypes were determined by PCR with genotype-specific primers targeting HCV core and NS5B regions. SVR was achieved in 91.1% of patients. Female sex, age > 70 years, and RAVs were significantly associated with non-SVR (p<0.01 for all). Propensity score-matching results among the patients without RAVs regarding sex, age, and fibrosis revealed that mixed HCV infection determined by HCV NS5B genotyping showed significantly lower SVR rates than 1B-mono infection (p=0.02). Female sex and RAVs were significant factors associated with treatment failure of this combination therapy for patients with HCV serogroup 1 infection. Mixed HCV infection other than 1B-mono infection would be useful for predicting treatment failure.

Metastatic ability and the epithelial-mesenchymal transition in induced cancer stem-like hepatoma cells.

Cancer stem cells (CSCs) are thought to play important roles in cancer malignancy. Previously, we successfully induced sphere cancer stem-like cells (CSLCs) from several cell lines and observed the property of chemoresistance. In the present study, we examined the metastatic potential of these induced CSLCs. Sphere cancer stem-like cells were induced from a human hepatoma cell line (SK-HEP-1) in a unique medium containing neural survival factor-1. Splenic injection of cells into immune-deficient mice was used to assess hematogenous liver metastasis. Transcriptomic strand-specific RNA-sequencing analysis, quantitative real-time PCR, and flow cytometry were carried out to examine the expression of epithelial-mesenchymal transition (EMT)-related genes. Splenic injection of CSLCs resulted in a significantly increased frequency of liver metastasis compared to parental cancer cells (P < .05). In CSLCs, a mesenchymal marker, Vimentin, and EMT-promoting transcription factors, Snail and Twist1, were upregulated compared to parental cells. Correspondingly, significant enrichment of the molecular signature of the EMT in CSLCs relative to parental cancer cells was shown (q < 0.01) by RNA-sequencing analysis. This analysis also revealed differential expression of CD44 isoforms between CSLCs and parental cancer cells. Increasing CD44 isoforms containing an extra exon were observed, and the standard CD44 isoform decreased in CSLCs compared to parental cells. Interestingly, another CD44 variant isoform encoding a short cytoplasmic tail was also upregulated in CSLCs (11.7-fold). Our induced CSLCs possess an increased liver metastatic potential in which promotion of the EMT and upregulation of CD44 variant isoforms, especially short-tail, were observed.

Bioisostere Identification by Determining the Amino Acid Binding Preferences of Common Chemical Fragments.

To assist in the structural optimization of hit/lead compounds during drug discovery, various computational approaches to identify potentially useful bioisosteric conversions have been reported. Here, the preference of chemical fragments to hydrogen bonds with specific amino acid residues was used to identify potential bioisosteric conversions. We first compiled a data set of chemical fragments frequently occurring in complex structures contained in the Protein Data Bank. We then used a computational approach to determine the amino acids to which these chemical fragments most frequently hydrogen bonded. The results of the frequency analysis were used to hierarchically cluster chemical fragments according to their amino acid preferences. The Euclid distance between amino acid preferences of chemical fragments for hydrogen bonding was then compared to MMP information in the ChEMBL database. To demonstrate the applicability of the approach for compound optimization, the similarity of amino acid preferences was used to identify known bioisosteric conversions of the epidermal growth factor receptor inhibitor gefitinib. The amino acid preference distance successfully detected bioisosteric fragments corresponding to the morpholine ring in gefitinib with a higher ROC score compared to those based on topological similarity of substituents and frequency of MMP in the ChEMBL database.

Unsupervised class labeling of diffuse lung diseases using frequent attribute patterns.

PURPOSE: For realizing computer-aided diagnosis (CAD) of computed tomography (CT) images, many pattern recognition methods have been applied to automatic classification of normal and abnormal opacities; however, for the learning of accurate classifier, a large number of images with correct labels are necessary. It is a very time-consuming and impractical task for radiologists to give correct labels for a large number of CT images. In this paper, to solve the above problem and realize an unsupervised class labeling mechanism without using correct labels, a new clustering algorithm for diffuse lung diseases using frequent attribute patterns is proposed. METHODS: A large number of frequently appeared patterns of opacities are extracted by a data mining algorithm named genetic network programming (GNP), and the extracted patterns are automatically distributed to several clusters using genetic algorithm (GA). In this paper, lung CT images are used to make clusters of normal and diffuse lung diseases. RESULTS: After executing the pattern extraction by GNP, 1,148 frequent attribute patterns were extracted; then, GA was executed to make clusters. This paper deals with making clusters of normal and five kinds of abnormal opacities (i.e., six-class problem), and then, the proposed method without using correct class labels in the training showed 47.7 % clustering accuracy. CONCLUSION: It is clarified that the proposed method can make clusters without using correct labels and has the potential to apply to CAD, reducing the time cost for labeling CT images.

Deformable image registration between pathological images and MR image via an optical macro image.

Computed tomography (CT) and magnetic resonance (MR) imaging have been widely used for visualizing the inside of the human body. However, in many cases, pathological diagnosis is conducted through a biopsy or resection of an organ to evaluate the condition of tissues as definitive diagnosis. To provide more advanced information onto CT or MR image, it is necessary to reveal the relationship between tissue information and image signals. We propose a registration scheme for a set of PT images of divided specimens and a 3D-MR image by reference to an optical macro image (OM image) captured by an optical camera. We conducted a fundamental study using a resected human brain after the death of a brain cancer patient. We constructed two kinds of registration processes using the OM image as the base for both registrations to make conversion parameters between the PT and MR images. The aligned PT images had shapes similar to the OM image. On the other hand, the extracted cross-sectional MR image was similar to the OM image. From these resultant conversion parameters, the corresponding region on the PT image could be searched and displayed when an arbitrary pixel on the MR image was selected. The relationship between the PT and MR images of the whole brain can be analyzed using the proposed method. We confirmed that same regions between the PT and MR images could be searched and displayed using resultant information obtained by the proposed method. In terms of the accuracy of proposed method, the TREs were 0.56±0.39mm and 0.87±0.42mm. We can analyze the relationship between tissue information and MR signals using the proposed method.

Development of a 2D Image Reconstruction and Viewing System for Histological Images from Multiple Tissue Blocks: Towards High-Resolution Whole-Organ 3D Histological Images.

High-resolution 3D histology image reconstruction of the whole brain organ starts from reconstructing the high-resolution 2D histology images of a brain slice. In this paper, we introduced a method to automatically align the histology images of thin tissue sections cut from the multiple paraffin-embedded tissue blocks of a brain slice. For this method, we employed template matching and incorporated an optimization technique to further improve the accuracy of the 2D reconstructed image. In the template matching, we used the gross image of the brain slice as a reference to the reconstructed 2D histology image of the slice, while in the optimization procedure, we utilized the Jaccard index as the metric of the reconstruction accuracy. The results of our experiment on the initial 3 different whole-brain tissue slices showed that while the method works, it is also constrained by tissue deformations introduced during the tissue processing and slicing. The size of the reconstructed high-resolution 2D histology image of a brain slice is huge, and designing an image viewer that makes particularly efficient use of the computing power of a standard computer used in our laboratories is of interest. We also present the initial implementation of our 2D image viewer system in this paper.

Alpha-fetoprotein before and after pegylated interferon therapy for predicting hepatocellular carcinoma development.

AIM: To investigate factors that accurately predict hepatocellular carcinoma (HCC) development after antiviral therapy in chronic hepatitis C (CHC) patients. METHODS: CHC patients who received pegylated interferon and ribavirin were enrolled in this cohort study that investigated the ability of alpha-fetoprotein (AFP) to predict HCC development after interferon (IFN) therapy. RESULTS: Of 1255 patients enrolled, 665 developed sustained virological response (SVR) during mean follow-up period of 5.4 years. HCC was occurred in 89 patients, and 20 SVR patients were included. Proportional hazard models showed that HCC occurred in SVR patients showing AFP ≥ 5 ng/mL before therapy and in non-SVR patients showing AFP ≥ 5 ng/mL before and 1 year after therapy besides older age, and low platelet counts. SVR patients showing AFP ≥ 5 ng/mL before therapy and no decrease in AFP to < 5 ng/mL 1 year after therapy had significantly higher HCC incidence than non-SVR patients showing AFP ≥ 5 ng/mL before therapy and decreased AFP (P = 0.043). AFP ≥ 5 ng/mL before therapy was significantly associated with low platelet counts and high values of alanine aminotransferase (ALT) in stepwise logistic regression analysis. After age, gender, platelet count, and ALT was matched by propensity score, significantly lower HCC incidence was shown in SVR patients showing AFP < 5 ng/mL before therapy than in those showing AFP ≥ 5 ng/mL. CONCLUSION: The criteria of AFP < 5 ng/mL before and 1 year after IFN therapy is a benefical predictor for HCC development in CHC patients.

Effectiveness of extending treatment duration in therapy with pegylated interferon and ribavirin for genotype 2 hepatitis C virus infection.

The effectiveness of extending treatment duration as response guided therapy was previously reported for chronic hepatitis C (CHC) genotype 1, but is still controversial for genotype 2. The present study is a retrospective cohort study to investigate the effectiveness of extending treatment duration in therapy with pegylated interferon and ribavirin for patients with CHC genotype 2 by focusing on the timing at which patients obtained undetectable HCV RNA. A total of 306 patients who obtained undetectable HCV RNA by week 24 of treatment and completed 24 weeks of treatment were enrolled. Rapid virological response (RVR) to standard therapy was achieved by 122 patients (51%), and 89% of them obtained sustained virological response (SVR), while 69% of non-RVR patients achieved SVR. Non-RVR patients with undetectable HCV RNA at week 8, and insufficient adherence＜80% pegylated interferon and ribavirin during the first 24 weeks, significantly improved their SVR rate by extended therapy. Among patients receiving extended therapy, drug adherences did not differ between SVR and non-SVR patients, indicating that extending treatment duration might compensate for insufficient antiviral effects due to insufficient drug adherences. This finding might be useful in creating a guideline for extending treatment duration for patients with CHC genotype 2.

Randomized trial of low-dose peginterferon α-2b plus low and escalating doses of ribavirin in older patients with chronic hepatitis C with high viral load genotype 1.

Elderly patients with chronic hepatitis C cannot tolerate standard combination therapy of peginterferon and ribavirin, which remains the backbone of therapy in many countries, including Japan. The efficacy and safety of low-dose peginterferon α-2b in combination with low and escalating doses of ribavirin in older patients with high viral load genotype 1 were investigated in this randomized controlled trial. Thirty-two patients (age ≥ 60 years) were randomized into standard (group 1) or low (group 2) doses of peginterferon α-2b in combination with low and escalating doses of ribavirin. Patients were evaluated for safety and efficacy of treatment. There was a higher virological response rate in group 1 than in group 2. However, the response in men was higher than in women in the early treatment phase and 24 weeks after treatment (P = 0.008). There was no significant difference between the two groups in the virological response rate in men and women. Completion of therapy was higher in group 2 than in group 1 (31% vs. 13%, P = 0.200). Dose modification of ribavirin was less frequent in group 2 than in group 1 (69% vs. 88%, P = 0.200). These data suggest that combination therapy with low-dose peginterferon plus low and escalating doses of ribavirin may be safer in older patients than that with standard dose peginterferon, without impairing the treatment response.