Predicting blood-brain barrier permeability of molecules with a large language model and machine learning.

Predicting the blood-brain barrier (BBB) permeability of small-molecule compounds using a novel artificial intelligence platform is necessary for drug discovery. Machine learning and a large language model on artificial intelligence (AI) tools improve the accuracy and shorten the time for new drug development. The primary goal of this research is to develop artificial intelligence (AI) computing models and novel deep learning architectures capable of predicting whether molecules can permeate the human blood-brain barrier (BBB). The in silico (computational) and in vitro (experimental) results were validated by the Natural Products Research Laboratories (NPRL) at China Medical University Hospital (CMUH). The transformer-based MegaMolBART was used as the simplified molecular input line entry system (SMILES) encoder with an XGBoost classifier as an in silico method to check if a molecule could cross through the BBB. We used Morgan or Circular fingerprints to apply the Morgan algorithm to a set of atomic invariants as a baseline encoder also with an XGBoost classifier to compare the results. BBB permeability was assessed in vitro using three-dimensional (3D) human BBB spheroids (human brain microvascular endothelial cells, brain vascular pericytes, and astrocytes). Using multiple BBB databases, the results of the final in silico transformer and XGBoost model achieved an area under the receiver operating characteristic curve of 0.88 on the held-out test dataset. Temozolomide (TMZ) and 21 randomly selected BBB permeable compounds (Pred scores = 1, indicating BBB-permeable) from the NPRL penetrated human BBB spheroid cells. No evidence suggests that ferulic acid or five BBB-impermeable compounds (Pred scores < 1.29423E-05, which designate compounds that pass through the human BBB) can pass through the spheroid cells of the BBB. Our validation of in vitro experiments indicated that the in silico prediction of small-molecule permeation in the BBB model is accurate. Transformer-based models like MegaMolBART, leveraging the SMILES representations of molecules, show great promise for applications in new drug discovery. These models have the potential to accelerate the development of novel targeted treatments for disorders of the central nervous system.

A Study to Evaluate Accuracy and Validity of the EFAI Computer-Aided Bone Age Diagnosis System Compared With Qualified Physicians.

Study Objectives: In previous research, we built a deep neural network model based on Inception-Resnet-v2 to predict bone age (EFAI-BAA). The primary objective of the study was to determine if the EFAI-BAA was substantially concordant with the qualified physicians in assessing bone ages. The secondary objective of the study was to determine if the EFAI-BAA was no different in the clinical rating (advanced, normal, or delayed) with the qualified physicians. Method: This was a retrospective study. The left-hand X-ray images of male subjects aged 3-16 years old and female subjects aged 2-15 years old were collected from China Medical University Hospital (CMUH) and Asia University Hospital (AUH) retrospectively since the trial began until the included image amount reached 368. This was a blinded study. The qualified physicians who ran, read, and interpreted the tests were blinded to the values assessed by the other qualified physicians and the EFAI-BAA. Results: The concordance correlation coefficient (CCC) between the EFAI-BAA (EFAI-BAA), the evaluation of bone age by physician in Kaohsiung Veterans General Hospital (KVGH), Taichung Veterans General Hospital (TVGH2), and in Taipei Tzu Chi Hospital (TZUCHI-TP) was 0.9828 (95% CI: 0.9790-0.9859, p-value = 0.6782), 0.9739 (95% CI: 0.9681-0.9786, p-value = 0.0202), and 0.9592 (95% CI: 0.9501-0.9666, p-value = 0.4855), respectively. Conclusion: There was a consistency of bone age assessment between the EFAI-BAA and each one of the three qualified physicians (CCC = 0.9). As the significant difference in the clinical rating was only found between the EFAI-BAA and the qualified physician in TVGH2, the performance of the EFAI-BAA was considered similar to the qualified physicians.

A Phase I, Open-Label, Dose-Escalation Study of the OX40 Agonist Ivuxolimab in Patients with Locally Advanced or Metastatic Cancers.

PURPOSE: Stimulation of effector T cells is an appealing immunotherapeutic approach in oncology. OX40 (CD134) is a costimulatory receptor expressed on activated CD4+ and CD8+ T cells. Induction of OX40 following antigen recognition results in enhanced T-cell activation, proliferation, and survival, and OX40 targeting shows therapeutic efficacy in preclinical studies. We report the monotherapy dose-escalation portion of a multicenter, phase I trial (NCT02315066) of ivuxolimab (PF-04518600), a fully human immunoglobulin G2 agonistic monoclonal antibody specific for human OX40. PATIENTS AND METHODS: Adult patients (N = 52) with selected locally advanced or metastatic cancers received ivuxolimab 0.01 to 10 mg/kg. Primary endpoints were safety and tolerability. Secondary/exploratory endpoints included preliminary assessment of antitumor activity and biomarker analyses. RESULTS: The most common all-causality adverse events were fatigue (46.2%), nausea (28.8%), and decreased appetite (25.0%). Of 31 treatment-related adverse events, 30 (96.8%) were grade ≤2. No dose-limiting toxicities occurred. Ivuxolimab exposure increased in a dose-proportionate manner from 0.3 to 10 mg/kg. Full peripheral blood target engagement occurred at ≥0.3 mg/kg. Three (5.8%) patients achieved a partial response, and disease control was achieved in 56% of patients. Increased CD4+ central memory T-cell proliferation and activation, and clonal expansion of CD4+ and CD8+ T cells in peripheral blood were observed at 0.1 to 3.0 mg/kg. Increased immune cell infiltrate and OX40 expression were evident in on-treatment tumor biopsies. CONCLUSIONS: Ivuxolimab was generally well tolerated with on-target immune activation at clinically relevant doses, showed preliminary antitumor activity, and may serve as a partner for combination studies.

Report of clinical bone age assessment using deep learning for an Asian population in Taiwan.

INTRODUCTION: A deep learning-based automatic bone age identification system (ABAIs) was introduced in medical imaging. This ABAIs enhanced accurate, consistent, and timely clinical diagnostics and enlightened research fields of deep learning and artificial intelligence (AI) in medical imaging. AIM: The goal of this study was to use the Deep Neural Network (DNN) model to assess bone age in months based on a database of pediatric left-hand radiographs. METHODS: The Inception Resnet V2 model with a Global Average Pooling layer to connect to a single fully connected layer with one neuron using the Rectified Linear Unit (ReLU) activation function consisted of the DNN model for bone age assessment (BAA) in this study. The medical data in each case contained posterior view of X-ray image of left hand, information of age, gender and weight, and clinical skeletal bone assessment. RESULTS: A database consisting of 8,061 hand radiographs with their gender and age (0-18 years) as the reference standard was used. The DNN model's accuracies on the testing set were 77.4%, 95.3%, 99.1% and 99.7% within 0.5, 1, 1.5 and 2 years of the ground truth respectively. The MAE for the study subjects was 0.33 and 0.25 year for male and female models, respectively. CONCLUSION: In this study, Inception Resnet V2 model was used for automatic interpretation of bone age. The convolutional neural network based on feature extraction has good performance in the bone age regression model, and further improves the accuracy and efficiency of image-based bone age evaluation. This system helps to greatly reduce the burden on clinical personnel.

Automatic detection and characterization of quantitative phase images of thalassemic red blood cells using a mask region-based convolutional neural network.

SIGNIFICANCE: Label-free quantitative phase imaging is a promising technique for the automatic detection of abnormal red blood cells (RBCs) in real time. Although deep-learning techniques can accurately detect abnormal RBCs from quantitative phase images efficiently, their applications in diagnostic testing are limited by the lack of transparency. More interpretable results such as morphological and biochemical characteristics of individual RBCs are highly desirable. AIM: An end-to-end deep-learning model was developed to efficiently discriminate thalassemic RBCs (tRBCs) from healthy RBCs (hRBCs) in quantitative phase images and segment RBCs for single-cell characterization. APPROACH: Two-dimensional quantitative phase images of hRBCs and tRBCs were acquired using digital holographic microscopy. A mask region-based convolutional neural network (Mask R-CNN) model was trained to discriminate tRBCs and segment individual RBCs. Characterization of tRBCs was achieved utilizing SHapley Additive exPlanation analysis and canonical correlation analysis on automatically segmented RBC phase images. RESULTS: The implemented model achieved 97.8% accuracy in detecting tRBCs. Phase-shift statistics showed the highest influence on the correct classification of tRBCs. Associations between the phase-shift features and three-dimensional morphological features were revealed. CONCLUSIONS: The implemented Mask R-CNN model accurately identified tRBCs and segmented RBCs to provide single-RBC characterization, which has the potential to aid clinical decision-making.

Radiomics features analysis of PET images in oropharyngeal and hypopharyngeal cancer.

This study used radiomics image analysis to examine the differences of texture feature values extracted from oropharyngeal and hypopharyngeal cancer positron emission tomography (PET) images on various tumor segmentations, and finds the proper and stable feature groups. A total of 80 oropharyngeal and hypopharyngeal cancer cases were retrospectively recruited. Radiomics method was applied to the PET image for the 80 oropharyngeal and hypopharyngeal cancer cases to extract texture features from various defined metabolic volumes. Kruskal-Wallis one-way analysis of variance method was used to test whether feature value difference exists between groups, which were grouped by stage, response to treatment, and recurrence. If there was a significant difference, the corresponding feature cutoff value was applied to the Kaplan-Meier estimator to estimate the survival functions. For the various defined metabolic volumes, there were 16 features that had significant differences between early (T1, T2) and late tumor stages (T3, T4). Five images and 2 textural features were found to be able to predict the tumor response and recurrence, respectively, with the areas under the receiver operating characteristic curves reaching 0.7. The histogram entropy was found to be a good predictor of overall survival (OS) and primary relapse-free survival (PRFS) of oropharyngeal and hypopharyngeal cancer patients. Textural features from PET images provide predictive and prognostic information in tumor staging, tumor response, recurrence, and have the potential to be a prognosticator for OS and PRFS in oropharyngeal and hypopharyngeal cancer.

Detection and classification the breast tumors using mask R-CNN on sonograms.

Breast cancer is one of the most harmful diseases for women with the highest morbidity. An efficient way to decrease its mortality is to diagnose cancer earlier by screening. Clinically, the best approach of screening for Asian women is ultrasound images combined with biopsies. However, biopsy is invasive and it gets incomprehensive information of the lesion. The aim of this study is to build a model for automatic detection, segmentation, and classification of breast lesions with ultrasound images. Based on deep learning, a technique using Mask regions with convolutional neural network was developed for lesion detection and differentiation between benign and malignant. The mean average precision was 0.75 for the detection and segmentation. The overall accuracy of benign/malignant classification was 85%. The proposed method provides a comprehensive and noninvasive way to detect and classify breast lesions.

Spirally configured cis-stilbene/fluorene hybrids as bipolar, organic sensitizers for solar cell applications.

Hybrids based on a dibenzosuberene core bearing a spiro-fluorene junction at the C-5 position and with amino donor and ß-thiophenyl-α-cyanoacrylic acid acceptor groups at C-3 and C-7, respectively, serve as new organic sensitizer materials for solar cell applications. Solar cell devices based on these materials show a conversion efficiency (η) of up to 6.1% (V(oc) = 697 mV, J(sc) = 12.2 mA cm(-2), FF = 0.72) under AM 1.5 G conditions. The best IPCE values exceed 75% within the 450-550 nm absorption range.

Interfacial rheology and structure of tiled graphene oxide sheets.

The hydrophilic nature of graphene oxide sheets can be tailored by varying the carbon to oxygen ratio. Depending on this ratio, the particles can be deposited at either a water-air or a water-oil interface. Upon compression of thus-created Langmuir monolayers, the sheets cover the entire interface, assembling into a strong, compact layer of tiled graphene oxide sheets. With further compression, the particle layer forms wrinkles that are reversible upon expansion, resembling the behavior of an elastic membrane. In the present work, we investigate under which conditions the structure and properties of the interfacial layer are such that free-standing films can be obtained. The interfacial rheological properties of these films are investigated using both compressional experiments and shear rheometry. The role of surface rheology in potential applications of such tiled films is explored. The rheological properties are shown to be responsible for the efficiency of such layers in stabilizing water-oil emulsions. Moreover, because of the mechanical integrity, large-area monolayers can be deposited by, for example, Langmuir-Blodgett techniques using aqueous subphases. These films can be turned into transparent conductive films upon subsequent chemical reduction.

Cytotoxicity of graphene oxide and graphene in human erythrocytes and skin fibroblasts.

Two-dimensional carbon-based nanomaterials, including graphene oxide and graphene, are potential candidates for biomedical applications such as sensors, cell labeling, bacterial inhibition, and drug delivery. Herein, we explore the biocompatibility of graphene-related materials with controlled physical and chemical properties. The size and extent of exfoliation of graphene oxide sheets was varied by sonication intensity and time. Graphene sheets were obtained from graphene oxide by a simple (hydrazine-free) hydrothermal route. The particle size, morphology, exfoliation extent, oxygen content, and surface charge of graphene oxide and graphene were characterized by wide-angle powder X-ray diffraction, atomic force microscopy, X-ray photoelectron spectroscopy, dynamic light scattering, and zeta-potential. One method of toxicity assessment was based on measurement of the efflux of hemoglobin from suspended red blood cells. At the smallest size, graphene oxide showed the greatest hemolytic activity, whereas aggregated graphene sheets exhibited the lowest hemolytic activity. Coating graphene oxide with chitosan nearly eliminated hemolytic activity. Together, these results demonstrate that particle size, particulate state, and oxygen content/surface charge of graphene have a strong impact on biological/toxicological responses to red blood cells. In addition, the cytotoxicity of graphene oxide and graphene sheets was investigated by measuring mitochondrial activity in adherent human skin fibroblasts using two assays. The methylthiazolyldiphenyl-tetrazolium bromide (MTT) assay, a typical nanotoxicity assay, fails to predict the toxicity of graphene oxide and graphene toxicity because of the spontaneous reduction of MTT by graphene and graphene oxide, resulting in a false positive signal. However, appropriate alternate assessments, using the water-soluble tetrazolium salt (WST-8), trypan blue exclusion, and reactive oxygen species assay reveal that the compacted graphene sheets are more damaging to mammalian fibroblasts than the less densely packed graphene oxide. Clearly, the toxicity of graphene and graphene oxide depends on the exposure environment (i.e., whether or not aggregation occurs) and mode of interaction with cells (i.e., suspension versus adherent cell types).

Aqueous only route toward graphene from graphite oxide.

We report a new, simple, hydrazine-free, high-yield method for producing single-layer graphene sheets. Graphene sheets were formed from graphite oxide by reduction with simple deionized water at 95 °C under atmospheric pressure. Over 65% of the sheets are single graphene layers; the average sheet diameter is 300 nm. We speculate that dehydration of graphene oxide is the main mechanism for oxygen reduction and transformation of C-C bonds from sp(3) to sp(2). The reduction appears to occur in large uniform interconnected oxygen-free patches so that despite the presence of residual oxygen the sp(2) carbon bonds formed on the sheets are sufficient to provide electronic properties comparable to reduced graphene sheets obtained using other methods.

A novel polyurethane-based root canal-obturation material and urethane-acrylate-based root canal sealer-part 2: evaluation of push-out bond strengths.

We have developed a visible-light curable urethane-acrylate/tripropylene glycol diacrylate (UA/TPGDA) oligomer to serve as a root canal sealer and a zinc oxide/thermoplastic polyurethane (ZnO/TPU) composite to serve as a root canal obturation material. The purpose of this study was to compare the push-out bond strengths of the following 8 groups of materials: (1) Tubliseal + gutta-percha (TB/GP); (2) Tubliseal + Resilon (TB/R); (3) Epiphany + gutta-percha (EP/GP); (4) Epiphany + Resilon (EP/R); (5) EndoREZ sealer + EndoREZ cone (ES/EC); (6) EndoREZ sealer + ZnO/TPU (ES/PU); (7) UA/TPGDA + EndoREZ cone (UA/EC); and (8) UA/TPGDA + ZnO/TPU (UA/PU). Eighty 1-mm-thick root slices prepared from extracted human permanent molars were randomly divided into 8 groups with 10 specimens in each group. Root slices were filled with the above obturation materials, and then push-out test was performed with a universal testing machine. The results showed that the UA/EC and UA/PU groups had significantly higher bond strengths than the other groups.

A novel polyurethane-based root canal-obturation material and urethane acrylate-based root canal sealer--part I: synthesis and evaluation of mechanical and thermal properties.

Resilon (RealSeal; SybronEndo, Orange, CA) has been developed as an alternative to gutta percha, but its advantages over gutta percha remain controversial. In this study, we developed a novel zinc oxide/thermoplastic polyurethane (ZnO/TPU) composite root canal-filling material and a visible-light curable urethane-acrylate/tripropylene glycol diacrylate (UA/TPGDA) root canal sealer. The mechanical and thermal properties of the ZnO/TPU composite were compared with those of gutta percha and Resilon. Results showed that the tensile strength and elastic modulus of the ZnO/TPU composite were markedly higher than those of gutta percha and Resilon. The melting points of all three materials were similar; however, the enthalpy change and specific heat of ZnO/TPU (9.4 J/g, 0.7 J/g degrees C) were close to those of gutta percha (10.9 J/g, 0.7 J/g degrees C) but lower than those of Resilon (28.9 J/g, 1.3 J/g degrees C). The results indicate that ZnO/TPU composite exhibits better mechanical strength than Resilon, and its combination with UA/TPGDA sealer has excellent potential to be used as a root canal-filling material.