Biopsy or Follow-up: AI Improves the Clinical Strategy of US BI-RADS 4A Breast Nodules Using a Convolutional Neural Network.

OBJECTIVES: To develop predictive nomograms based on clinical and ultrasound features and to improve the clinical strategy for US BI-RADS 4A lesions. METHODS: Patients with US BI-RADS 4A lesions from 3 hospitals between January 2016 and June 2020 were retrospectively included. Clinical and ultrasound features were extracted to establish nomograms CE (based on clinical experience) and DL (based on deep-learning algorithm). The performances of nomograms were evaluated by receiver operator characteristic curves, calibration curves and decision curves. Diagnostic performances with DL of radiologists were analyzed. RESULTS: 1616 patients from 2 hospitals were randomly divided into training and internal validation cohorts at a ratio of 7:3. Hundred patients from another hospital made up external validation cohort. DL achieved more optimized AUCs than CE (internal validation: 0.916 vs. 0.863, P < .01; external validation: 0.884 vs. 0.776, P = .05). The sensitivities of DL were higher than CE (internal validation: 81.03% vs. 72.41%, P = .044; external validation: 93.75% vs. 81.25%, P = .4795) without losing specificity (internal validation: 84.91% vs. 86.47%, P = .353; external validation: 69.14% vs. 71.60%, P = .789). Decision curves indicated DL adds more clinical net benefit. With DL's assistance, both radiologists achieved higher AUCs (0.712 vs. 0.801; 0.547 vs. 0.800), improved specificities (70.93% vs. 74.42%, P < .001; 59.3% vs. 81.4%, P = .004), and decreased unnecessary biopsy rates by 6.7% and 24%. CONCLUSION: DL was developed to discriminate US BI-RADS 4A lesions with a higher diagnostic power and more clinical net benefit than CE. Using DL may guide clinicians to make precise clinical decisions and avoid overtreatment of benign lesions.

An integrated model incorporating deep learning, hand-crafted radiomics and clinical and US features to diagnose central lymph node metastasis in patients with papillary thyroid cancer.

OBJECTIVE: To evaluate the value of an integrated model incorporating deep learning (DL), hand-crafted radiomics and clinical and US imaging features for diagnosing central lymph node metastasis (CLNM) in patients with papillary thyroid cancer (PTC). METHODS: This retrospective study reviewed 613 patients with clinicopathologically confirmed PTC from two institutions. The DL model and hand-crafted radiomics model were developed using primary lesion images and then integrated with clinical and US features selected by multivariate analysis to generate an integrated model. The performance was compared with junior and senior radiologists on the independent test set. SHapley Additive exPlanations (SHAP) plot and Gradient-weighted Class Activation Mapping (Grad-CAM) were used for the visualized explanation of the model. RESULTS: The integrated model yielded the best performance with an AUC of 0.841. surpassing that of the hand-crafted radiomics model (0.706, p < 0.001) and the DL model (0.819, p = 0.26). Compared to junior and senior radiologists, the integrated model reduced the missed CLNM rate from 57.89% and 44.74-27.63%, and decreased the rate of unnecessary central lymph node dissection (CLND) from 29.87% and 27.27-18.18%, respectively. SHAP analysis revealed that the DL features played a primary role in the diagnosis of CLNM, while clinical and US features (such as extrathyroidal extension, tumour size, age, gender, and multifocality) provided additional support. Grad-CAM indicated that the model exhibited a stronger focus on thyroid capsule in patients with CLNM. CONCLUSION: Integrated model can effectively decrease the incidence of missed CLNM and unnecessary CLND. The application of the integrated model can help improve the acceptance of AI-assisted US diagnosis among radiologists.

Biosynthesis of 10-Hydroxy-2-decenoic Acid through a One-Step Whole-Cell Catalysis.

10-Hydroxy-2-decenoic acid (10-HDA) is an important component of royal jelly, known for its antimicrobial, anti-inflammatory, blood pressure-lowering, and antiradiation effects. Currently, 10-HDA biosynthesis is limited by the substrate selectivity of acyl-coenzyme A dehydrogenase, which restricts the technique to a two-step process. This study aimed to develop an efficient and simplified method for synthesizing 10-HDA. In this study, ACOX from Candida tropicalis 1798, which catalyzes 10-hydroxydecanoyl coenzyme A desaturation for 10-HDA synthesis, was isolated and heterologously coexpressed with FadE, Macs, YdiI, and CYP in Escherichia coli/SK after knocking out FadB, FadJ, and FadR genes. The engineered E. coli/AKS strain achieved a 49.8% conversion of decanoic acid to 10-HDA. CYP expression was improved through ultraviolet mutagenesis and high-throughput screening, increased substrate conversion to 75.6%, and the synthesis of 10-HDA was increased to 0.628 g/L in 10 h. This is the highest conversion rate and product concentration achieved in the shortest time to date. This study provides a simple and efficient method for 10-HDA biosynthesis and offers an effective method for developing strains with high product yields.

Palladium-Catalyzed ß-C(sp3)-H Bond Arylation of Tertiary Aldehydes Facilitated by 2-Pyridone Ligands.

2-Pyridone ligand-facilitated palladium-catalyzed direct C-H bond functionalization via the transient directing group strategy has become an attractive topic. Here, we report a Pd-catalyzed direct ß-C(sp3)-H arylation reaction of tertiary aliphatic aldehydes by using an α-amino acid as a transient directing group in combination with a 2-pyridone ligand.

Diagnostic performance of deep learning in ultrasound diagnosis of breast cancer: a systematic review.

Deep learning (DL) has been widely investigated in breast ultrasound (US) for distinguishing between benign and malignant breast masses. This systematic review of test diagnosis aims to examine the accuracy of DL, compared to human readers, for the diagnosis of breast cancer in the US under clinical settings. Our literature search included records from databases including PubMed, Embase, Scopus, and Cochrane Library. Test accuracy outcomes were synthesized to compare the diagnostic performance of DL and human readers as well as to evaluate the assistive role of DL to human readers. A total of 16 studies involving 9238 female participants were included. There were no prospective studies comparing the test accuracy of DL versus human readers in clinical workflows. Diagnostic test results varied across the included studies. In 14 studies employing standalone DL systems, DL showed significantly lower sensitivities in 5 studies with comparable specificities and outperformed human readers at higher specificities in another 4 studies; in the remaining studies, DL models and human readers showed equivalent test outcomes. In 12 studies that assessed assistive DL systems, no studies proved the assistive role of DL in the overall diagnostic performance of human readers. Current evidence is insufficient to conclude that DL outperforms human readers or enhances the accuracy of diagnostic breast US in a clinical setting. Standardization of study methodologies is required to improve the reproducibility and generalizability of DL research, which will aid in clinical translation and application.

A bimodal nomogram as an adjunct tool to reduce unnecessary breast biopsy following discordant ultrasonic and mammographic BI-RADS assessment.

OBJECTIVE: To develop a bimodal nomogram to reduce unnecessary biopsies in breast lesions with discordant ultrasound (US) and mammography (MG) Breast Imaging Reporting and Data System (BI-RADS) assessments. METHODS: This retrospective study enrolled 706 women following opportunistic screening or diagnosis with discordant US and MG BI-RADS assessments (where one assessed a lesion as BI-RADS 4 or 5, while the other assessed the same lesion as BI-RADS 0, 2, or 3) from two medical centres between June 2019 and June 2021. Univariable and multivariable logistic regression analyses were used to develop the nomogram. DeLong's and McNemar's tests were used to assess the model's performance. RESULTS: Age, MG features (margin, shape, and density in masses, suspicious calcifications, and architectural distortion), and US features (margin and shape in masses as well as calcifications) were independent risk factors for breast cancer. The nomogram obtained an area under the curve of 0.87 (95% confidence interval (CI), 0.83-0.91), 0.91 (95% CI, 0.87 - 0.96), and 0.92 (95% CI, 0.86-0.98) in the training, internal validation, and external testing samples, respectively, and demonstrated consistency in calibration curves. Coupling the nomogram with US reduced unnecessary biopsies from 74 to 44% and the missed malignancies rate from 13 to 2%. Similarly, coupling with MG reduced missed malignancies from 20 to 6%, and 63% of patients avoided unnecessary biopsies. Interobserver agreement between US and MG increased from - 0.708 (poor agreement) to 0.700 (substantial agreement) with the nomogram. CONCLUSION: When US and MG BI-RADS assessments are discordant, incorporating the nomogram may improve the diagnostic accuracy, avoid unnecessary breast biopsies, and minimise missed diagnoses. CLINICAL RELEVANCE STATEMENT: The nomogram developed in this study could be used as a computer program to assist radiologists with detecting breast cancer and ensuring more precise management and improved treatment decisions for breast lesions with discordant assessments in clinical practice. KEY POINTS: • Coupling the nomogram with US and mammography improves the detection of breast cancers without the risk of unnecessary biopsy or missed malignancies. • The nomogram increases mammography and US interobserver agreement and enhances the consistency of decision-making. • The nomogram has the potential to be a computer program to assist radiologists in identifying breast cancer and making optimal decisions.

Flow-cytometric cell sorting coupled with UV mutagenesis for improving pectin lyase expression.

Introduction: Alkaline pectin lyase is an important enzyme with a wide range of applications in industrial production, It has been widely used in many important fields such as fruit juice processing and extraction, the dyeing and processing of cotton and linen textiles, degumming plant fibers, environmental industrial wastewater treatment, and pulp and paper production. PGLA-rep4 was previously generated as a modified alkaline pectin lyase with high specific activity at pH 11.0°C and 70°C. However, the pre-constructed high-activity pectin lyase expression strains are still difficult to apply in industrial production due to their limited enzymatic activity. We hope to solve these problems by combining modern breeding techniques with high-throughput equipment to rapidly screen alkaline pectin lyase with higher enzymatic activity and lower cost. Methods: We fused the genes encoding PGLA-rep4 and fluorescent protein egfp using a flexible linker peptide and ligated them into a temperature-sensitive plasmid, pKD46. The constructed screening plasmid pKD46-PGLA-rep4-egfp was then transformed into an expression host and screened via flow-cytometric cell sorting coupled with UV mutagenesis. Results: Following mutagenesis, primary screening, and secondary screening, the high-expression strain, named Escherichia coli BL21/1G3, was obtained. The screening plasmid pKD46-PGLA-rep4-egfp was eliminated, and the original expression plasmid pET28a-PGLA-rep4 was then retransformed into the mutant strains. After induction and fermentation, pectin lyase activity in E. coli BL21/1G3 was significantly increased (1.37-fold relative to that in the parental E. coli BL21/PGLA-rep4 strain, p < 0.001), and the highest activity was 230, 240 U/mL at 144 h. Genome sequencing revealed that genes encoding ribonuclease E (RNase E) and diadenosine tetraphosphatase (ApaH) of E. coli BL21/1G3 were mutated compared to the sequence in the original E. coli BL21 (DE3) strain, which could be associated with increased enzyme expression. Discussion: Our work provides an effective method for the construction of strains expressing pectin lyase at high levels.

Dual nanoenzymes loaded hollow mesoporous organotantalum nanospheres for chemo-radio sensitization.

Chemo-radiotherapy has been extensively used in clinics, displaying substantial advantages in treatment and prognosis. Stimuli-responsive biodegradable nanoagents that can achieve not only delivery and controlled release of chemotherapeutics, but also hypoxia alleviation to enhance chemoradiotherapy therefore has tremendous potential. Herein, glutathione (GSH)-responsive, biodegradable, doxorubicin-carrying hollow mesoporous organotantalum nanospheres modified with Au and Pt dual nanoenzymes (HMOTP@Pt@Au@Dox) were constructed for chemo-radio sensitization. Degradation of HMOTP@Pt@Au@Dox can be self-activated through GSH stimulation and on-demand release packaged Dox owing to the disulfide bond in the hybrid framework of organotantalum nanospheres. Au and Pt nanoenzymes triggered cascade catalytic reactions that could alleviate hypoxia by utilizing ß-d-glucose and H2O2, thereby sensitizing ROS-based chemoradiotherapy with synergistic starving therapy. Given the radiosensitization of high-Z elements (Ta, Pt, Au), nanoenzymes induced cascade catalytic reaction for hypoxia relief, and the depletion of the predominant antioxidant GSH, desirable tumor suppression could be achieved both in vitro and in vivo, indicating that HMOTP@Pt@Au@Dox is a promising nanoagent to boost chemo-radiotherapy.

Molecular Engineering of Fully Conjugated sp2 Carbon-Linked Polymers for High-Efficiency Photocatalytic Hydrogen Evolution.

The diverse nature of organic precursors offers a versatile platform for precisely tailoring the electronic properties of semiconducting polymers. In this study, three fully conjugated sp2 carbon-linked polymers have been designed and synthesized for photocatalytic hydrogen evolution under visible-light illumination, by copolymerizing different C3 -symmetric aromatic aldehydes as knots with the 1,4-phenylene diacetonitrile (PDAN) linker through a C=C condensation reaction. The hydrogen evolution (HER) is achieved at a maximum rate of 30.2 mmol g-1 h-1 over a polymer based on 2,4,6-triphenyl-1,3,5-triazine units linked by cyano-substituted phenylene, with an apparent quantum yield (AQY) of 7.20 % at 420 nm. Increasing the degree of conjugation and planarity not only extends visible-light absorption, but also stabilizes the fully conjugated sp2 -carbon-linked donor-acceptor (D-A) polymer. Incorporating additional electron-withdrawing triazine units into the D-A polymer to form multiple electron donors and acceptors can greatly promote exciton separation and charge transfer, thus significantly enhancing the photocatalytic activity.

Hot π-Electron Tunneling of Metal-Insulator-COF Nanostructures for Efficient Hydrogen Production.

A metal-insulator-semiconductor (MIS) photosystem based on covalent organic framework (COF) semiconductors was designed for robust and efficient hydrogen evolution under visible-light irradiation. A maximal H2 evolution rate of 8.42 mmol h-1 g-1 and a turnover frequency of 789.5 h-1 were achieved by using a MIS photosystem prepared by electrostatic self-assembly of polyvinylpyrrolidone (PVP) insulator-capped Pt nanoparticles (NPs) with the hydrophilic imine-linked TP-COFs having =C=O-H-N= hydrogen-bonding groups. The hot π-electrons in the photoexcited n-type TP-COF semiconductors can be efficiently extracted and tunneled to Pt NPs across an ultrathin PVP insulating layer to reduce protons to H2 . Compared to the Schottky-type counterparts, the COF-based MIS photosystems give a 32-fold-enhanced carrier efficiency, attributed to the combined enhancement of photoexcitation rate, charge separation, and oxidation rate of holes accumulated in the valence band of the TP-COF semiconductor.

Photocatalytic and antibacterial properties of medical-grade PVC material coated with TiO2 film.

The TiO(2) film was coated on poly vinyl chloride (PVC) surface by dip-coating process from TiO(2)-PVC-THF suspension. The morphology and crystal structure of the as-synthesized samples were characterized by SEM and XRD. The photocatalytic properties were measured by the photodegradation reaction of RhB and the anti-adhesion and anti-bacteria for Escherichia coli. The results show that the resultant TiO(2) film is well-conglutinated on PVC surface and has the same crystal structure as the original TiO(2) powder. The TiO(2)/PVC shows excellent photocatalytic activity for the degradation of aqueous RhB and the activity increases with increasing reaction time and tends toward stable after accumulative illumination for 11.5 h. The TiO(2) film shows good bacterial anti-adhesion activity following photo-activation and sterilization property under UV irradiation. The E. coli can be killed completely after UV irradiation for 1.5 h.