Optimization of the stroke hospital selection strategy and the distribution of endovascular thrombectomy resources.

Nowadays, emergency medical technicians (EMTs) decide to send a suspected stroke patient to a primary stroke center (PSC) or to an endovascular thrombectomy (EVT)-capable hospital, based on the Cincinnati Prehospital Stroke Scale (CPSS) and the number of symptoms a patient presents at the scene. Based on existing studies, the patient is likely to have a better functional outcome after three months if the time between the onset of symptoms and receiving EVT treatment is shorter. However, if an acute ischemic stroke (AIS) patient with large vessel occlusion (LVO) is first sent to a PSC, and then needs to be transferred to an EVT-capable hospital, the time to get definitive treatment is significantly increased. For this purpose, We formulate an integer programming model to minimize the expected time to receive a definitive treatment for stroke patients. We then use real-world data to verify the validity of the model. Also, we expand our model to find the optimal redistribution and centralization of EVT resources. It will enable therapeutic teams to increase their experience and skills more efficiently within a short period of time.

Structure-Guided Discovery of PD-1/PD-L1 Interaction Inhibitors: Peptide Design, Screening, and Optimization via Computation-Aided Phage Display Engineering.

Cancer immunotherapy harnesses the immune system to combat tumors and has emerged as a major cancer treatment modality. The PD-1/PD-L1 immune checkpoint modulates interactions between tumor cells and T cells and has been extensively targeted in cancer immunotherapy. However, the monoclonal antibodies known to target this immune checkpoint have considerable side effects, and novel PD-1/PD-L1 inhibitors are therefore required. Herein, a peptide inhibitor to disrupt PD-1/PD-L1 interactions was designed through structure-driven phage display engineering coupled to computational modification and optimization. BetaPb, a novel peptide library constructed by using the known structure of PD-1/PD-L, was used to develop inhibitors against the immune checkpoint, and specific peptides with high affinity toward PD-1 were screened through enzyme-linked immunosorbent assays, homogeneous time-resolved fluorescence, and biolayer interferometry. A potential inhibitor, B8, was preliminarily screened through biopanning. The binding affinity of B8 toward PD-1 was confirmed through computation-aided optimization. Assessment of B8 variants (B8.1, B8.2, B8.3, B8.4, and B8.5) demonstrated their attenuation of PD-1/PD-L1 interactions. B8.4 exhibited the strongest attenuation efficiency at a half-maximal effective concentration of 0.1 µM and the strongest binding affinity to PD-1 (equilibrium dissociation constant = 0.1 µM). B8.4 outperformed the known PD-1/PD-L1 interaction inhibitor PL120131 in disrupting PD-1/PD-L1 interactions, revealing that B8.4 has remarkable potential for modification to yield an antitumor agent. This study provides valuable information for the future development of peptide-based drugs, therapeutics, and immunotherapies for cancer.

An auristatin-based antibody-drug conjugate targeting EphA2 in pancreatic cancer treatment.

Pancreatic adenocarcinoma, a highly aggressive form of cancer with a poor prognosis, necessitates the development of innovative treatment strategies. Our prior research showcased the growth-inhibiting effects of the anti-EphA2 antibody drug hSD5 on pancreatic cancer tumors. This antibody targets and induces the degradation of the EphA2 receptor while also prompting the antibody's internalization. A deeper dive into the hSD5 Fab crystallographic structure and docking studies revealed that hSD5's CDRH3 drives the primary interaction between hSD5 and the EphA2 active site. In this study, we developed a novel antibody-drug conjugate (ADC)-the auristatin-based hSD5-vedotin specifically targeting EphA2 in pancreatic cancer cells. This ADC aims at the tumor-specific antigen EphA2, triggering endocytosis and releasing the conjugated payload molecule Monomethyl auristatin E (MMAE), amplifying the tumor-killing effect. Upon cellular entry, hSD5-vedotin demonstrated an impressive tumor-killing response, inhibiting tumor cell growth and promoting apoptosis even at lower antibody concentrations. In a pancreatic cancer xenograft animal model, hSD5-vedotin showcased the potential to suppress tumor growth entirely. Notably, potential immune resistance responses were also observed in recurrent pancreatic cancer tumors. Our empirical results underscore the possibility of developing hSD5-vedotin further, which we anticipate will have a broader and more potent therapeutic impact on pancreatic cancer and other EphA2-related cancers.

Effectiveness of anti-erythropoietin producing Hepatocellular receptor Type-A2 antibody in pancreatic cancer treatment.

Erythropoietin-producing hepatocyte receptor type A2 (EphA2) is a tyrosine kinase that binds to ephrins (e.g., ephrin-A1) to initiate bidirectional signaling between cells. The binding of EphA2 and ephrin-A1 leads to the inhibition of Ras-MAPK activity and tumor growth. During tumorigenesis, the normal interaction between EphA2 and ephrin-A1 is hindered, which leads to the overexpression of EphA2 and induces cancer. The overexpression of EphA2 has been identified as a notable tumor marker in diagnosing and treating pancreatic cancer. In this study, we used phage display to isolate specific antibodies against the active site of EphA2 by using a discontinuous recombinant epitope for immunization. The therapeutic efficacy and inhibition mechanism of the generated antibody against pancreatic cancer was validated and clarified. The generated antibodies were bound to the conformational epitope of endogenous EphA2 on cancer cells, thus inducing cellular endocytosis and causing EphA2 degradation. Molecule signals pAKT, pERK, pFAK, and pSTAT3 were weakened, inhibiting the proliferation and migration of pancreatic cancer cells. The humanized antibody hSD5 could effectively inhibit the growth of the xenograft pancreatic cancer tumor cells BxPc-3 and Mia PaCa-2 in mice, respectively. When antibody hSD5 was administered with gemcitabine, significantly improved effects on tumor growth inhibition were observed. Based on the efficacy of the IgG hSD5 antibodies, clinical administration of the hSD5 antibodies is likely to suppress tumors in patients with pancreatic cancer and abnormal activation or overexpression of EphA2 signaling.

Single-chain fragment antibody disrupting the EphA4 function as a therapeutic drug for gastric cancer.

Studies have shown that the high expression of EphA4 in gastric cancer tissues may correlate with unfavorable clinical pathological characteristics. Therefore, EphA4 may be an effective target for treating gastric cancer in addition to HER-2/neu. In this study, generated scFv S3 can bind endogenous EphA4 of gastric cancer cells and has significant membrane staining. Additionally, scFv S3 binding to EphA4 inhibits the growth and migration of cancer cells and the growth induction that ephrinA1 generates in gastric cancer cells. We found that EphA4 molecules may degrade through antibody treatment of cells, and the increase in LAMP1 and LAMP2 indicates that lysosome is involved in the degradation. The scFv S3 administration leads to the signals pAKT, pERK, and pSTAT3 decrease in cancer cells. The xenograft model of HER-2/neu low expressing gastric cancer cell SNU-16 exhibits better therapeutic effects by scFv S3 than trastuzumab scFv. The scFv S3 administration in vivo can degrade EphA4 molecules in tumor tissues, decreasing Ki67 and increasing cleaved C3 molecule expression. Furthermore, we identified and validated that scFv S3 generates essential ionic bonding with R162 on EphA4. The antibody may provide effective treatment for patients with gastric cancer and abnormal activation or overexpression of EphA4 signaling.

Deep Learning Can Predict Bevacizumab Therapeutic Effect and Microsatellite Instability Directly from Histology in Epithelial Ovarian Cancer.

Epithelial ovarian cancer (EOC) remains a significant cause of mortality among gynecologic cancers, with the majority of cases being diagnosed at an advanced stage. Before targeted therapies were available, EOC treatment relied largely on debulking surgery and platinum-based chemotherapy. Vascular endothelial growth factors have been identified as inducing tumor angiogenesis. According to several clinical trials, anti-vascular endothelial growth factor-targeted therapy with bevacizumab was effective in all phases of EOC treatment. However, there are currently no biomarkers accessible for regular therapeutic use despite the importance of patient selection. Microsatellite instability (MSI), caused by a deficiency of the DNA mismatch repair system, is a molecular abnormality observed in EOC associated with Lynch syndrome. Recent evidence suggests that angiogenesis and MSI are interconnected. Developing predictive biomarkers, which enable the selection of patients who might benefit from bevacizumab-targeted therapy or immunotherapy, is critical for realizing personalized precision medicine. In this study, we developed 2 improved deep learning methods that eliminate the need for laborious detailed image-wise annotations by pathologists and compared them with 3 state-of-the-art methods to not only predict the efficacy of bevacizumab in patients with EOC using mismatch repair protein immunostained tissue microarrays but also predict MSI status directly from histopathologic images. In prediction of therapeutic outcomes, the 2 proposed methods achieved excellent performance by obtaining the highest mean sensitivity and specificity score using MSH2 or MSH6 markers and outperformed 3 state-of-the-art deep learning methods. Moreover, both statistical analysis results, using Cox proportional hazards model analysis and Kaplan-Meier progression-free survival analysis, confirm that the 2 proposed methods successfully differentiate patients with positive therapeutic effects and lower cancer recurrence rates from patients experiencing disease progression after treatment (P < .01). In prediction of MSI status directly from histopathology images, our proposed method also achieved a decent performance in terms of mean sensitivity and specificity score even for imbalanced data sets for both internal validation using tissue microarrays from the local hospital and external validation using whole section slides from The Cancer Genome Atlas archive.

Detection of ERBB2 and CEN17 signals in fluorescent in situ hybridization and dual in situ hybridization for guiding breast cancer HER2 target therapy.

The overexpression of the human epidermal growth factor receptor 2 (HER2) is a predictive biomarker in therapeutic effects for metastatic breast cancer. Accurate HER2 testing is critical for determining the most suitable treatment for patients. Fluorescent in situ hybridization (FISH) and dual in situ hybridization (DISH) have been recognized as FDA-approved methods to determine HER2 overexpression. However, analysis of HER2 overexpression is challenging. Firstly, the boundaries of cells are often unclear and blurry, with large variations in cell shapes and signals, making it challenging to identify the precise areas of HER2-related cells. Secondly, the use of sparsely labeled data, where some unlabeled HER2-related cells are classified as background, can significantly confuse fully supervised AI learning and result in unsatisfactory model outcomes. In this study, we present a weakly supervised Cascade R-CNN (W-CRCNN) model to automatically detect HER2 overexpression in HER2 DISH and FISH images acquired from clinical breast cancer samples. The experimental results demonstrate that the proposed W-CRCNN achieves excellent results in identification of HER2 amplification in three datasets, including two DISH datasets and a FISH dataset. For the FISH dataset, the proposed W-CRCNN achieves an accuracy of 0.970±0.022, precision of 0.974±0.028, recall of 0.917±0.065, F1-score of 0.943±0.042 and Jaccard Index of 0.899±0.073. For DISH datasets, the proposed W-CRCNN achieves an accuracy of 0.971±0.024, precision of 0.969±0.015, recall of 0.925±0.020, F1-score of 0.947±0.036 and Jaccard Index of 0.884±0.103 for dataset 1, and an accuracy of 0.978±0.011, precision of 0.975±0.011, recall of 0.918±0.038, F1-score of 0.946±0.030 and Jaccard Index of 0.884±0.052 for dataset 2, respectively. In comparison with the benchmark methods, the proposed W-CRCNN significantly outperforms all the benchmark approaches in identification of HER2 overexpression in FISH and DISH datasets (p<0.05). With the high degree of accuracy, precision and recall , the results show that the proposed method in DISH analysis for assessment of HER2 overexpression in breast cancer patients has significant potential to assist precision medicine.

CW-NET for multitype cell detection and classification in bone marrow examination and mitotic figure examination.

MOTIVATION: Bone marrow (BM) examination is one of the most important indicators in diagnosing hematologic disorders and is typically performed under the microscope via oil-immersion objective lens with a total 100× objective magnification. On the other hand, mitotic detection and identification is critical not only for accurate cancer diagnosis and grading but also for predicting therapy success and survival. Fully automated BM examination and mitotic figure examination from whole-slide images is highly demanded but challenging and poorly explored. First, the complexity and poor reproducibility of microscopic image examination are due to the cell type diversity, delicate intralineage discrepancy within the multitype cell maturation process, cells overlapping, lipid interference and stain variation. Second, manual annotation on whole-slide images is tedious, laborious and subject to intraobserver variability, which causes the supervised information restricted to limited, easily identifiable and scattered cells annotated by humans. Third, when the training data are sparsely labeled, many unlabeled objects of interest are wrongly defined as background, which severely confuses AI learners. RESULTS: This article presents an efficient and fully automatic CW-Net approach to address the three issues mentioned above and demonstrates its superior performance on both BM examination and mitotic figure examination. The experimental results demonstrate the robustness and generalizability of the proposed CW-Net on a large BM WSI dataset with 16 456 annotated cells of 19 BM cell types and a large-scale WSI dataset for mitotic figure assessment with 262 481 annotated cells of five cell types. AVAILABILITY AND IMPLEMENTATION: An online web-based system of the proposed method has been created for demonstration (see https://youtu.be/MRMR25Mls1A).

The Blockade of Mitogen-Activated Protein Kinase 14 Activation by Marine Natural Product Crassolide Triggers ICD in Tumor Cells and Stimulates Anti-Tumor Immunity.

Immunogenic cell death (ICD) refers to a type of cell death that stimulates immune responses. It is characterized by the surface exposure of damage-associated molecular patterns (DAMPs), which can facilitate the uptake of antigens by dendritic cells (DCs) and stimulate DC activation, resulting in T cell immunity. The activation of immune responses through ICD has been proposed as a promising approach for cancer immunotherapy. The marine natural product crassolide, a cembranolide isolated from the Formosan soft coral Lobophytum michaelae, has been shown to have cytotoxic effects on cancer cells. In this study, we investigated the effects of crassolide on the induction of ICD, the expression of immune checkpoint molecules and cell adhesion molecules, as well as tumor growth in a murine 4T1 mammary carcinoma model. Immunofluorescence staining for DAMP ectolocalization, Western blotting for protein expression and Z'-LYTE kinase assay for kinase activity were performed. The results showed that crassolide significantly increased ICD and slightly decreased the expression level of CD24 on the surface of murine mammary carcinoma cells. An orthotopic tumor engraftment of 4T1 carcinoma cells indicated that crassolide-treated tumor cell lysates stimulate anti-tumor immunity against tumor growth. Crassolide was also found to be a blocker of mitogen-activated protein kinase 14 activation. This study highlights the immunotherapeutic effects of crassolide on the activation of anticancer immune responses and suggests the potential clinical use of crassolide as a novel treatment for breast cancer.

Development of anti-aflatoxin B1 nanobodies from a novel mutagenesis-derived synthetic library for traditional Chinese medicine and foods safety testing.

BACKGROUND: The main commercially available methods for detecting small molecules of mycotoxins in traditional Chinese medicine (TCM) and functional foods are enzyme-linked immunosorbent assay and mass spectrometry. Regarding the development of diagnostic antibody reagents, effective methods for the rapid preparation of specific monoclonal antibodies are inadequate. METHODS: In this study, a novel synthetic phage-displayed nanobody Golden Glove (SynaGG) library with a glove-like cavity configuration was established using phage display technology in synthetic biology. We applied this unique SynaGG library on the small molecule aflatoxin B1 (AFB1), which has strong hepatotoxicity, to isolate specific nanobodies with high affinity for AFB1. RESULT: These nanobodies exhibit no cross-reactivity with the hapten methotrexate, which is recognized by the original antibody template. By binding to AFB1, two nanobodies can neutralize AFB1-induced hepatocyte growth inhibition. Using molecular docking, we found that the unique non-hypervariable complementarity-determining region 4 (CDR4) loop region of the nanobody was involved in the interaction with AFB1. Specifically, the CDR4's positively charged amino acid arginine directed the binding interaction between the nanobody and AFB1. We then rationally optimized the interaction between AFB1 and the nanobody by mutating serine at position 2 into valine. The binding affinity of the nanobody to AFB1 was effectively improved, and this result supported the use of molecular structure simulation for antibody optimization. CONCLUSION: In summary, this study revealed that the novel SynaGG library, which was constructed through computer-aided design, can be used to isolate nanobodies that specifically bind to small molecules. The results of this study could facilitate the development of nanobody materials to detect small molecules for the rapid screening of TCM materials and foods in the future.

Effect of using G-FAST to recognize emergent large vessel occlusion: A city-wide community experience.

BACKGROUND/PURPOSE: A prehospital bypass strategy was suggested for large vessel occlusion. This study aimed to evaluate the effect of a bypass strategy using the gaze-face-arm-speech-time test (G-FAST) implemented in a metropolitan community. METHODS: Pre-notified patients with positive Cincinnati Prehospital Stroke Scale and symptom onset <3 h from July 2016 to December 2017 (pre-intervention period) and those with positive G-FAST and symptom onset <6 h from July 2019 to December 2020 (intervention period) were included. Patients aged <20 years and those with missing in-hospital data were excluded. The primary outcomes were the rates of receiving endovascular thrombectomy (EVT) and intravenous thrombolysis (IVT). The secondary outcomes were total prehospital time, door-to-computed tomography (CT) time, door-to-needle (DTN) time, and door-to-puncture (DTP) time. RESULTS: We included 802 and 695 pre-notified patients from the pre-intervention and intervention periods, respectively. The characteristics of the patients in the two periods were similar. In the primary outcomes, pre-notified patients during the intervention period showed higher rates of receiving EVT (4.49% vs. 15.25%, p < 0.001) and IVT (15.34% vs. 21.58%, p = 0.002). In the secondary outcomes, pre-notified patients during intervention period had longer total prehospital time (mean 23.38 vs 25.23 min, p < 0.001), longer door-to-CT time (median 10 vs 11 min, p < 0.001), longer DTN time (median 53 vs 54.5 min, p < 0.001) but shorter DTP time (median 141 vs 139.5 min, p < 0.001). CONCLUSION: The prehospital bypass strategy with G-FAST showed benefits for stroke patients.

Interpretable attention-based deep learning ensemble for personalized ovarian cancer treatment without manual annotations.

Inhibition of pathological angiogenesis has become one of the first FDA approved targeted therapies widely tested in anti-cancer treatment, i.e. VEGF-targeting monoclonal antibody bevacizumab, in combination with chemotherapy for frontline and maintenance therapy for women with newly diagnosed ovarian cancer. Identification of the best predictive biomarkers of bevacizumab response is necessary in order to select patients most likely to benefit from this therapy. Hence, this study investigates the protein expression patterns on immunohistochemical whole slide images of three angiogenesis related proteins, including Vascular endothelial growth factor, Angiopoietin 2 and Pyruvate kinase isoform M2, and develops an interpretable and annotation-free attention based deep learning ensemble framework to predict the bevacizumab therapeutic effect on patients with epithelial ovarian cancer or peritoneal serous papillary carcinoma using tissue microarrays (TMAs). In evaluation with five-fold cross validation, the proposed ensemble model using the protein expressions of both Pyruvate kinase isoform M2 and Angiopoietin 2 achieves a notably high F-score (0.99±0.02), accuracy (0.99±0.03), precision (0.99±0.02), recall (0.99±0.02) and AUC (1.00±0). Kaplan-Meier progression free survival analysis confirms that the proposed ensemble is able to identify patients in the predictive therapeutic sensitive group with low cancer recurrence (p<0.001), and the Cox proportional hazards model analysis further confirms the above statement (p=0.012). In conclusion, the experimental results demonstrate that the proposed ensemble model using the protein expressions of both Pyruvate kinase isoform M2 and Angiopoietin 2 can assist treatment planning of bevacizumab targeted therapy for patients with ovarian cancer.

Annotation-Free Deep Learning-Based Prediction of Thyroid Molecular Cancer Biomarker BRAF (V600E) from Cytological Slides.

Thyroid cancer is the most common endocrine cancer. Papillary thyroid cancer (PTC) is the most prevalent form of malignancy among all thyroid cancers arising from follicular cells. Fine needle aspiration cytology (FNAC) is a non-invasive method regarded as the most cost-effective and accurate diagnostic method of choice in diagnosing PTC. Identification of BRAF (V600E) mutation in thyroid neoplasia may be beneficial because it is specific for malignancy, implies a worse prognosis, and is the target for selective BRAF inhibitors. To the authors' best knowledge, this is the first automated precision oncology framework effectively predict BRAF (V600E) immunostaining result in thyroidectomy specimen directly from Papanicolaou-stained thyroid fine-needle aspiration cytology and ThinPrep cytological slides, which is helpful for novel targeted therapies and prognosis prediction. The proposed deep learning (DL) framework is evaluated on a dataset of 118 whole slide images. The results show that the proposed DL-based technique achieves an accuracy of 87%, a precision of 94%, a sensitivity of 91%, a specificity of 71% and a mean of sensitivity and specificity at 81% and outperformed three state-of-the-art deep learning approaches. This study demonstrates the feasibility of DL-based prediction of critical molecular features in cytological slides, which not only aid in accurate diagnosis but also provide useful information in guiding clinical decision-making in patients with thyroid cancer. With the accumulation of data and the continuous advancement of technology, the performance of DL systems is expected to be improved in the near future. Therefore, we expect that DL can provide a cost-effective and time-effective alternative tool for patients in the era of precision oncology.

Production and characterization of single-chain variable fragment antibodies targeting the breast cancer tumor marker nectin-4.

Background: Nectin-4 is a novel biomarker overexpressed in various types of cancer, including breast cancer, in which it has been associated with poor prognosis. Current literature suggests that nectin-4 has a role in cancer progression and may have prognostic and therapeutic implications. The present study aims to produce nectin-4-specific single-chain variable fragment (scFv) antibodies and evaluate their applications in breast cancer cell lines and clinical specimens. Methods: We generated recombinant nectin-4 ectodomain fragments as immunogens to immunize chickens and the chickens' immunoglobulin genes were amplified for construction of anti-nectin-4 scFv libraries using phage display. The binding capacities of the selected clones were evaluated with the recombinant nectin-4 fragments, breast cancer cell lines, and paraffin-embedded tissue sections using various laboratory approaches. The binding affinity and in silico docking profile were also characterized. Results: We have selected two clones (S21 and L4) from the libraries with superior binding capacity. S21 yielded higher signals when used as the primry antibody for western blot analysis and flow cytometry, whereas clone L4 generated cleaner and stronger signals in immunofluorescence and immunohistochemistry staining. In addition, both scFvs could diminish attachment-free cell aggregation of nectin-4-positive breast cancer cells. As results from ELISA indicated that L4 bound more efficiently to fixed nectin-4 ectodomain, molecular docking analysis was further performed and demonstrated that L4 possesses multiple polar contacts with nectin-4 and diversity in interacting residues. Conclusion: Overall, the nectin-4-specific scFvs could recognize nectin-4 expressed by breast cancer cells and have the merit of being further explored for potential diagnostic and therapeutic applications.

Deep Learning Using Endobronchial-Ultrasound-Guided Transbronchial Needle Aspiration Image to Improve the Overall Diagnostic Yield of Sampling Mediastinal Lymphadenopathy.

Lung cancer is the biggest cause of cancer-related death worldwide. An accurate nodal staging is critical for the determination of treatment strategy for lung cancer patients. Endobronchial-ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) has revolutionized the field of pulmonology and is considered to be extremely sensitive, specific, and secure for lung cancer staging through rapid on-site evaluation (ROSE), but manual visual inspection on the entire slide of EBUS smears is challenging, time consuming, and worse, subjective, on a large interobserver scale. To satisfy ROSE's needs, a rapid, automated, and accurate diagnosis system using EBUS-TBNA whole-slide images (WSIs) is highly desired to improve diagnosis accuracy and speed, minimize workload and labor costs, and ensure reproducibility. We present a fast, efficient, and fully automatic deep-convolutional-neural-network-based system for advanced lung cancer staging on gigapixel EBUS-TBNA cytological WSIs. Each WSI was converted into a patch-based hierarchical structure and examined by the proposed deep convolutional neural network, generating the segmentation of metastatic lesions in EBUS-TBNA WSIs. To the best of the authors' knowledge, this is the first research on fully automated enlarged mediastinal lymph node analysis using EBUS-TBNA cytological WSIs. We evaluated the robustness of the proposed framework on a dataset of 122 WSIs, and the proposed method achieved a high precision of 93.4%, sensitivity of 89.8%, DSC of 82.2%, and IoU of 83.2% for the first experiment (37.7% training and 62.3% testing) and a high precision of 91.8 ± 1.2, sensitivity of 96.3 ± 0.8, DSC of 94.0 ± 1.0, and IoU of 88.7 ± 1.8 for the second experiment using a three-fold cross-validation, respectively. Furthermore, the proposed method significantly outperformed the three state-of-the-art baseline models, including U-Net, SegNet, and FCN, in terms of precision, sensitivity, DSC, and Jaccard index, based on Fisher's least significant difference (LSD) test (p<0.001). For a computational time comparison on a WSI, the proposed method was 2.5 times faster than U-Net, 2.3 times faster than SegNet, and 3.4 times faster than FCN, using a single GeForce GTX 1080 Ti, respectively. With its high precision and sensitivity, the proposed method demonstrated that it manifested the potential to reduce the workload of pathologists in their routine clinical practice.

Fast cross-staining alignment of gigapixel whole slide images with application to prostate cancer and breast cancer analysis.

Joint analysis of multiple protein expressions and tissue morphology patterns is important for disease diagnosis, treatment planning, and drug development, requiring cross-staining alignment of multiple immunohistochemical and histopathological slides. However, cross-staining alignment of enormous gigapixel whole slide images (WSIs) at single cell precision is difficult. Apart from gigantic data dimensions of WSIs, there are large variations on the cell appearance and tissue morphology across different staining together with morphological deformations caused by slide preparation. The goal of this study is to build an image registration framework for cross-staining alignment of gigapixel WSIs of histopathological and immunohistochemical microscopic slides and assess its clinical applicability. To the authors' best knowledge, this is the first study to perform real time fully automatic cross staining alignment of WSIs with 40× and 20× objective magnification. The proposed WSI registration framework consists of a rapid global image registration module, a real time interactive field of view (FOV) localization model and a real time propagated multi-level image registration module. In this study, the proposed method is evaluated on two kinds of cancer datasets from two hospitals using different digital scanners, including a dual staining breast cancer data set with 43 hematoxylin and eosin (H&E) WSIs and 43 immunohistochemical (IHC) CK(AE1/AE3) WSIs, and a triple staining prostate cancer data set containing 30 H&E WSIs, 30 IHC CK18 WSIs, and 30 IHC HMCK WSIs. In evaluation, the registration performance is measured by not only registration accuracy but also computational time. The results show that the proposed method achieves high accuracy of 0.833 ± 0.0674 for the triple-staining prostate cancer data set and 0.931 ± 0.0455 for the dual-staining breast cancer data set, respectively, and takes only 4.34 s per WSI registration on average. In addition, for 30.23% data, the proposed method takes less than 1 s for WSI registration. In comparison with the benchmark methods, the proposed method demonstrates superior performance in registration accuracy and computational time, which has great potentials for assisting medical doctors to identify cancerous tissues and determine the cancer stage in clinical practice.

Weakly supervised deep learning for prediction of treatment effectiveness on ovarian cancer from histopathology images.

Despite the progress made during the last two decades in the surgery and chemotherapy of ovarian cancer, more than 70 % of advanced patients are with recurrent cancer and decease. Surgical debulking of tumors following chemotherapy is the conventional treatment for advanced carcinoma, but patients with such treatment remain at great risk for recurrence and developing drug resistance, and only about 30 % of the women affected will be cured. Bevacizumab is a humanized monoclonal antibody, which blocks VEGF signaling in cancer, inhibits angiogenesis and causes tumor shrinkage, and has been recently approved by FDA as a monotherapy for advanced ovarian cancer in combination with chemotherapy. Considering the cost, potential toxicity, and finding that only a portion of patients will benefit from these drugs, the identification of new predictive method for the treatment of ovarian cancer remains an urgent unmet medical need. In this study, we develop weakly supervised deep learning approaches to accurately predict therapeutic effect for bevacizumab of ovarian cancer patients from histopathological hematoxylin and eosin stained whole slide images, without any pathologist-provided locally annotated regions. To the authors' best knowledge, this is the first model demonstrated to be effective for prediction of the therapeutic effect of patients with epithelial ovarian cancer to bevacizumab. Quantitative evaluation of a whole section dataset shows that the proposed method achieves high accuracy, 0.882 ± 0.06; precision, 0.921 ± 0.04, recall, 0.912 ± 0.03; F-measure, 0.917 ± 0.07 using 5-fold cross validation and outperforms two state-of-the art deep learning approaches Coudray et al. (2018), Campanella et al. (2019). For an independent TMA testing set, the three proposed methods obtain promising results with high recall (sensitivity) 0.946, 0.893 and 0.964, respectively. The results suggest that the proposed method could be useful for guiding treatment by assisting in filtering out patients without positive therapeutic response to suffer from further treatments while keeping patients with positive response in the treatment process. Furthermore, according to the statistical analysis of the Cox Proportional Hazards Model, patients who were predicted to be invalid by the proposed model had a very high risk of cancer recurrence (hazard ratio = 13.727) than patients predicted to be effective with statistical signifcance (p < 0.05).

A Weakly Supervised Deep Learning Method for Guiding Ovarian Cancer Treatment and Identifying an Effective Biomarker.

Ovarian cancer is a common malignant gynecological disease. Molecular target therapy, i.e., antiangiogenesis with bevacizumab, was found to be effective in some patients of epithelial ovarian cancer (EOC). Although careful patient selection is essential, there are currently no biomarkers available for routine therapeutic usage. To the authors' best knowledge, this is the first automated precision oncology framework to effectively identify and select EOC and peritoneal serous papillary carcinoma (PSPC) patients with positive therapeutic effect. From March 2013 to January 2021, we have a database, containing four kinds of immunohistochemical tissue samples, including AIM2, c3, C5 and NLRP3, from patients diagnosed with EOC and PSPC and treated with bevacizumab in a hospital-based retrospective study. We developed a hybrid deep learning framework and weakly supervised deep learning models for each potential biomarker, and the experimental results show that the proposed model in combination with AIM2 achieves high accuracy 0.92, recall 0.97, F-measure 0.93 and AUC 0.97 for the first experiment (66% training and 34%testing) and high accuracy 0.86 ± 0.07, precision 0.9 ± 0.07, recall 0.85 ± 0.06, F-measure 0.87 ± 0.06 and AUC 0.91 ± 0.05 for the second experiment using five-fold cross validation, respectively. Both Kaplan-Meier PFS analysis and Cox proportional hazards model analysis further confirmed that the proposed AIM2-DL model is able to distinguish patients gaining positive therapeutic effects with low cancer recurrence from patients with disease progression after treatment (p < 0.005).