Testing Precision Limits of Neural Network-Based Quality Control Metrics in High-Throughput Digital Microscopy.

OBJECTIVE: Digital microscopy is used to monitor particulates such as protein aggregates within biopharmaceutical products. The images that result encode a wealth of information that is underutilized in pharmaceutical process monitoring. For example, images of particles in protein drug products typically are analyzed only to obtain particle counts and size distributions, even though the images also reflect particle characteristics such as shape and refractive index. Multiple groups have demonstrated that convolutional neural networks (CNNs) can extract information from images of protein aggregates allowing assignment of the likely stress at the "root-cause" of aggregation. A practical limitation of previous CNN-based approaches is that the potential aggregation-inducing stresses must be known a priori, disallowing identification of particles produced by unknown stresses. METHODS: We demonstrate an expanded CNN analysis of flow imaging microscopy (FIM) images incorporating judiciously chosen particle standards within a recently proposed "fingerprinting algorithm" (Biotechnol. & Bioeng. (2020) 117:3322) that allows detection of particles formed by unknown root-causes. We focus on ethylene tetrafluoroethylene (ETFE) microparticles as standard surrogates for protein aggregates. We quantify the sensitivity of the new algorithm to experimental parameters such as microscope focus and solution refractive index changes, and explore how FIM sample noise affects statistical testing procedures. RESULTS & CONCLUSIONS: Applied to real-world microscopy images of protein aggregates, the algorithm reproducibly detects complex, distinguishing "textural features" of particles that are not easily described by standard morphological measurements. This offers promise for quality control applications and for detecting shifts in protein aggregate populations due to stresses resulting from unknown process upsets.

Severity-onset prediction of COVID-19 via artificial-intelligence analysis of multivariate factors.

Progression to a severe condition remains a major risk factor for the COVID-19 mortality. Robust models that predict the onset of severe COVID-19 are urgently required to support sensitive decisions regarding patients and their treatments. In this study, we developed a multivariate survival model based on early-stage CT images and other physiological indicators and biomarkers using artificial-intelligence analysis to assess the risk of severe COVID-19 onset. We retrospectively enrolled 338 adult patients admitted to a hospital in China (severity rate, 31.9%; mortality rate, 0.9%). The physiological and pathological characteristics of the patients with severe and non-severe outcomes were compared. Age, body mass index, fever symptoms upon admission, coexisting hypertension, and diabetes were the risk factors for severe progression. Compared with the non-severe group, the severe group demonstrated abnormalities in biomarkers indicating organ function, inflammatory responses, blood oxygen, and coagulation function at an early stage. In addition, by integrating the intuitive CT images, the multivariable survival model showed significantly improved performance in predicting the onset of severe disease (mean time-dependent area under the curve = 0.880). Multivariate survival models based on early-stage CT images and other physiological indicators and biomarkers have shown high potential for predicting the onset of severe COVID-19.

Artificial intelligence-powered spatial analysis of tumor-infiltrating lymphocytes as a biomarker in locally advanced unresectable thymic epithelial neoplasm: A single-center, retrospective, longitudinal cohort study.

BACKGROUND: Thymic epithelial tumors (TET) are rare malignancies and lack well-defined biomarkers for neoadjuvant therapy. This study aimed to evaluate the clinical utility of artificial intelligence (AI)-powered tumor-infiltrating lymphocyte (TIL) analysis in TET. METHODS: Patients initially diagnosed with unresectable thymoma or thymic carcinoma who underwent neoadjuvant therapy between January 2004 and December 2021 formed our study population. Hematoxylin and eosin-stained sections from the initial biopsy and surgery were analyzed using an AI-powered spatial TIL analyzer. Intratumoral TIL (iTIL) and stromal TIL (sTIL) were quantified and their immune phenotype (IP) was identified. RESULTS: Thirty-five patients were included in this study. The proportion of patients with partial response to neoadjuvant therapy was higher in the group with nondesert IP in preneoadjuvant biopsy (63.6% vs. 17.6%, p = 0.038). A significant increase in both iTIL (median 22.18/mm2 vs. 340.69/mm2 , p < 0.001) and sTIL (median 175.19/mm2 vs. 531.02/mm2 , p = 0.004) was observed after neoadjuvant therapy. Patients with higher iTIL (>147/mm2 ) exhibited longer disease-free survival (median, 29 months vs. 12 months, p = 0.009) and overall survival (OS) (median, 62 months vs. 45 months, p = 0.002). Patients with higher sTIL (>232.1/mm2 ) exhibited longer OS (median 62 months vs. 30 months, p = 0.021). CONCLUSIONS: Nondesert IP in initial biopsy was associated with a better response to neoadjuvant therapy. Increased infiltration of both iTIL and sTIL in surgical specimens were associated with longer OS in patients with TET who underwent resection followed by neoadjuvant therapy.

Artificial intelligence-powered spatial analysis of tumor-infiltrating lymphocytes as a predictive biomarker for axitinib in adenoid cystic carcinoma.

BACKGROUND: This study analyzed the predictive value of artificial intelligence (AI)-powered tumor-infiltrating lymphocyte (TIL) analysis in recurrent or metastatic (R/M) adenoid cystic carcinoma (ACC) treated with axitinib. METHODS: Patients from a multicenter, prospective phase II trial evaluating axitinib efficacy in R/M ACC were included in this study. H&E whole-side images of archival tumor tissues were analyzed by Lunit SCOPE IO, an AI-powered spatial TIL analyzer. RESULTS: Twenty-seven patients were included in the analysis. The best response was stable disease, and the median progression-free survival (PFS) was 11.1 months (95% CI, 9.2-13.7 months). Median TIL densities in the cancer and surrounding stroma were 25.8/mm2 (IQR, 8.3-73.0) and 180.4/mm2 (IQR, 69.6-342.8), respectively. Patients with stromal TIL density >342.5/mm2 exhibited longer PFS (p = 0.012). CONCLUSIONS: Cancer and stromal area TIL infiltration were generally low in R/M ACC. Higher stromal TIL infiltration was associated with a longer PFS with axitinib treatment.

Palliative Care Is Proven.

Palliative care was officially recognized by the World Health Organization in 1990 as a distinct specialty dedicated to relieving suffering and improving quality of life for patients with serious illnesses. Journal of Palliative Medicine (JPM) was founded in 1997 in response to the need for a scientifically rigorous peer-reviewed journal to advance the field. In our first quarter of the century, JPM has become a leading global peer-reviewed scientific journal. What is the way forward? We engaged with this question in two ways. First, we utilized artificial intelligence techniques to analyze the trends of the articles published in JPM for the past two decades to discern key topic themes. Second, we applied human intelligence by convening seven panels of experts to discuss current topics of interest to the field as a separate strategy for discerning the future. Taken together, the way forward is clear. The field of palliative care has become broader and more subspecialized than anyone ever imagined at the beginning. The expansion of new knowledge has accelerated in all directions from its origins in the end-of-life care of patients with cancer. Although implementation science is of paramount importance, the barriers to implementation of this growing body of specialized knowledge lie not just with the need for more science. Some of the barriers lie within our field. The way forward requires confidence in what we know and the establishment of new collaborations outside of our field, including with people outside of traditional health care.

Radiomics and Artificial Intelligence Can Predict Malignancy of Solitary Pulmonary Nodules in the Elderly.

Solitary pulmonary nodules (SPNs) are a diagnostic and therapeutic challenge for thoracic surgeons. Although such lesions are usually benign, the risk of malignancy remains significant, particularly in elderly patients, who represent a large segment of the affected population. Surgical treatment in this subset, which usually presents several comorbidities, requires careful evaluation, especially when pre-operative biopsy is not feasible and comorbidities may jeopardize the outcome. Radiomics and artificial intelligence (AI) are progressively being applied in predicting malignancy in suspicious nodules and assisting the decision-making process. In this study, we analyzed features of the radiomic images of 71 patients with SPN aged more than 75 years (median 79, IQR 76-81) who had undergone upfront pulmonary resection based on CT and PET-CT findings. Three different machine learning algorithms were applied-functional tree, Rep Tree and J48. Histology was malignant in 64.8% of nodules and the best predictive value was achieved by the J48 model (AUC 0.9). The use of AI analysis of radiomic features may be applied to the decision-making process in elderly frail patients with suspicious SPNs to minimize the false positive rate and reduce the incidence of unnecessary surgery.

Artificial Intelligence Analysis of Outdoor Sports Risk Self-Assessment on Insurance Psychology.

The development potential of China's medical insurance market is huge, and the research on medical insurance demand has always been the focus of academic discussions. As a result, the discipline of behavioral economics is derived, which aims to explain the decision-making behavior of individual insurance consumption. Among them, the focus of this study was to investigate the influence of individual psychological characteristics and cognitive level on insurance behavior under the difference of reference points. This paper combined behavioral insurance, actuarial mathematics and the econometrics knowledge system, comprehensive theoretical analysis, and empirical tests and analyzed the impact mechanism of individual frame effect on medical insurance demand under different reference points at multiple levels. At the same time, based on the risk self-assessment of outdoor sports, the artificial intelligence of insurance psychology was analyzed. Based on the correlation vector machine algorithm and the theoretical basis combined with the dual perspective of insurance products, the expected utility model was established under the "guarantee framework", and the prospect theoretical model was established under the "profit and loss framework". The framing effect was used to measure the relative size of "guarantee utility" and "profit and loss utility", and a high-insurance-rate model and a low-insurance-rate model were established. The theoretical model analysis found that under the high insurance rate, because the "profit and loss utility" is positive, the size of the individual frame effect is positively correlated with the willingness to insure. Under the low insurance rate, because the "profit and loss utility" is negative, the size of the individual frame effect is negatively correlated with the willingness to insure. The research results of this paper show that insurance is an important beginning of insurance consumption behavior, which includes the complex mentality and emotion of consumers on insurance activities. The insurance demand of policyholders is formed by the joint action of external and internal incentives. Many factors such as income level and education level play an important role in insurance consumption decision making.

Using artificial intelligence to analyze emotion and facial action units following facial rejuvenation surgery.

INTRODUCTION: There remains a lack of standards in facial rejuvenation procedures, which may be attributed to the subjective measures used to determine surgical outcomes and success. The aim of this study was to evaluate the use of machine learning technology, i.e. FaceReader™, to objectively measure facial rejuvenation surgery outcomes. METHODS: Using a retrospective study design, we enrolled a cohort of patients undergoing high SMAS facelift with/without additional procedures during a one-year interval. The predictor variable was surgery done (pre- vs. postoperative). The outcome variables were 28 facial action units, happiness, and sadness emotions, detected by FaceReader™. Appropriate statistics were calculated at α = 0.05. RESULTS: The sample comprised of 15 patients (11 females, 15 Caucasians, mean age of 55.7 years). There was an average increase in detected happy emotion from 1.03% to 13.17% (p>0.01). Conversely, the average angry emotion detected decreased from 14.66% to 0.63% (p<0.05). There were no other distinct action unit patterns across the operation. CONCLUSION: Despite a small sample size, the results of this study suggest that FaceReader™ can be used as an objective outcome assessment tool in patients undergoing high SMAS facelift with/without its adjuncts.

Integrating Network Pharmacology, Transcriptome and Artificial Intelligence for Investigating Into the Effect and Mechanism of Ning Fei Ping Xue Decoction Against the Acute Respiratory Distress Syndrome.

Acute respiratory distress syndrome (ARDS) is a high-mortality disease and lacks effective pharmacotherapy. A traditional Chinese medicine (TCM) formula, Ning Fei Ping Xue (NFPX) decoction, was demonstrated to play a critical role in alleviating inflammatory responses of the lung. However, its therapeutic effectiveness in ARDS and active compounds, targets, and molecular mechanisms remain to be elucidated. The present study investigates the effects of NFPX decoction on ARDS mice induced by lipopolysaccharides (LPS). The results revealed that NFPX alleviated lung edema evaluated by lung ultrasound, decreased lung wet/Dry ratio, the total cell numbers of bronchoalveolar lavage fluid (BALF), and IL-1ß, IL-6, and TNF-α levels in BALF and serum, and ameliorated lung pathology in a dose-dependent manner. Subsequently, UPLC-HRMS was performed to establish the compounds of NFPX. A total of 150 compounds in NFPX were characterized. Moreover, integrating network pharmacology approach and transcriptional profiling of lung tissues were performed to predict the underlying mechanism. 37 active components and 77 targets were screened out, and a herbs-compounds-targets network was constructed. Differentially expressed genes (DEGs) were identified from LPS-treated mice compared with LPS combined with NFPX mice. GO, KEGG, and artificial intelligence analysis indicated that NFPX might act on various drug targets. At last, potential targets, HRAS, SMAD4, and AMPK, were validated by qRT-PCR in ARDS murine model. In conclusion, we prove the efficacy of NFPX decoction in the treatment of ARDS. Furthermore, integrating network pharmacology, transcriptome, and artificial intelligence analysis contributes to illustrating the molecular mechanism of NFPX decoction on ARDS.