Application of Artificial Intelligence in Pathology: Trends and Challenges.

Given the recent success of artificial intelligence (AI) in computer vision applications, many pathologists anticipate that AI will be able to assist them in a variety of digital pathology tasks. Simultaneously, tremendous advancements in deep learning have enabled a synergy with artificial intelligence (AI), allowing for image-based diagnosis on the background of digital pathology. There are efforts for developing AI-based tools to save pathologists time and eliminate errors. Here, we describe the elements in the development of computational pathology (CPATH), its applicability to AI development, and the challenges it faces, such as algorithm validation and interpretability, computing systems, reimbursement, ethics, and regulations. Furthermore, we present an overview of novel AI-based approaches that could be integrated into pathology laboratory workflows.

Therapeutic Applications of the CRISPR-Cas System.

The clustered regularly interspaced palindromic repeat (CRISPR)-Cas system has revolutionized genetic engineering due to its simplicity, stability, and precision since its discovery. This technology is utilized in a variety of fields, from basic research in medicine and biology to medical diagnosis and treatment, and its potential is unbounded as new methods are developed. The review focused on medical applications and discussed the most recent treatment trends and limitations, with an emphasis on CRISPR-based therapeutics for infectious disease, oncology, and genetic disease, as well as CRISPR-based diagnostics, screening, immunotherapy, and cell therapy. Given its promising results, the successful implementation of the CRISPR-Cas system in clinical practice will require further investigation into its therapeutic applications.

Reduced Pain by Mind-Body Intervention Correlates with Improvement of Shoulder Function in People with Shoulder Pain: A Randomized Controlled Trial.

Meditation and acupressure-like stimulations have been shown to relieve pain. The aim of this study was to determine whether a short bout of mind-body intervention combined with meditation and acupressure-like stimulation was able to alleviate shoulder pain and improve its function in a short time window. Sixty-five adults with shoulder pain were recruited and randomly classified into two groups. One group participated in an intervention which consisted of acupressure-like stimulation and meditation over a 5 min period. The other group was instructed to rest during this time. A visual analog scale (VAS) pain score and objective constant scores were measured before and after intervention to determine shoulder pain and range of motion (ROM), respectively. A two-way repeated measures analysis of variance with Bonferroni correction and a regression analysis were performed. VAS pain, objective constant score, flexion, abduction, and external rotation score showed significant interactions between time and group. The pain intensity was significantly reduced, while flexion and abduction were significantly improved, in the experimental group compared to the control group, after the intervention. In addition, the change of flexion negatively correlated with the change of pain intensity in the experimental group, but not in the control group. These results show that a short-term application of mind-body intervention significantly alleviates shoulder pain and improves shoulder movement, suggesting its potential use as a therapy for people with shoulder pain.

Diagnostic Accuracy of the Deep Learning Model for the Detection of ST Elevation Myocardial Infarction on Electrocardiogram.

We aimed to measure the diagnostic accuracy of the deep learning model (DLM) for ST-elevation myocardial infarction (STEMI) on a 12-lead electrocardiogram (ECG) according to culprit artery sorts. From January 2017 to December 2019, we recruited patients with STEMI who received more than one stent insertion for culprit artery occlusion. The DLM was trained with STEMI and normal sinus rhythm ECG for external validation. The primary outcome was the diagnostic accuracy of DLM for STEMI according to the three different culprit arteries. The outcomes were measured using the area under the receiver operating characteristic curve (AUROC), sensitivity (SEN), and specificity (SPE) using the Youden index. A total of 60,157 ECGs were obtained. These included 117 STEMI-ECGs and 60,040 normal sinus rhythm ECGs. When using DLM, the AUROC for overall STEMI was 0.998 (0.996-0.999) with SEN 97.4% (95.7-100) and SPE 99.2% (98.1-99.4). There were no significant differences in diagnostic accuracy within the three culprit arteries. The baseline wanders in false positive cases (83.7%, 345/412) significantly interfered with the accurate interpretation of ST elevation on an ECG. DLM showed high diagnostic accuracy for STEMI detection, regardless of the type of culprit artery. The baseline wanders of the ECGs could affect the misinterpretation of DLM.

Self-Supervised Learning with Electrocardiogram Delineation for Arrhythmia Detection.

Electrocardiogram (ECG) signals convey immense information that, when properly processed, can be used to diagnose various health conditions including arrhythmia and heart failure. Deep learning algorithms have been successfully applied to medical diagnosis, but existing methods heavily rely on abundant high-quality annotations which are expensive. Self-supervised learning (SSL) circumvents this annotation cost by pre-training deep neural networks (DNNs) on auxiliary tasks that do not require manual annotation. Despite its imminent need, SSL applications to ECG classification remain under-explored. In this work, we propose an SSL algorithm based on ECG delineation and show its effectiveness for arrhythmia classification. Our experiments demonstrate not only how the proposed algorithm enhances the DNN's performance across various datasets and fractions of labeled data, but also how features learnt via pre-training on one dataset can be trans-ferred when fine-tuned on a different dataset.

Genetic Variations within and between Blue Crab (Portunus trituberculatus) Groups.

The five oligonucleotide primers (oligo-primers) turned out a total of 335 fragments (FMs) (52.9%) in the blue crab (Portunus trituberculatus) group alpha and 298 FMs (47.1%) in the crab group beta, with the FM scales range varying from 100 bp to 2,000 bp. The highest band-sharing (BS) value (0.907) was found between individual's no. 19 and no. 20 within the blue crab group beta. Parties in the blue crab group beta (0.601±0.017) had higher BS rates than did parties from the crab group alpha (0.563±0.017) (p<0.05). The polar dendrogram got by the five oligo-primers points out two genetic extents: bundle I (BLUECRAB 01, 03, 04, 05, 06, 08, and 10) and bundle II (BLUECRAB 02, 07, 09. 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, and 22). The OPD-01 primer revealed 22 loci shared by all the examples of the as FMs of 1,000 bp. The oligo-primer OPA-05 made unique loci shared to each group (ULSEG), almost 400 bp and 500 bp, individually, in blue crab group beta. The remaining oligo-primers did not reveal any loci shared by the two crab groups (LSTG). The average number of ULSEG was diverse and 1.6-fold higher in the crab group beta than in the crab group alpha.

Automated Detection of Acute Myocardial Infarction Using Asynchronous Electrocardiogram Signals-Preview of Implementing Artificial Intelligence With Multichannel Electrocardiographs Obtained From Smartwatches: Retrospective Study.

BACKGROUND: When using a smartwatch to obtain electrocardiogram (ECG) signals from multiple leads, the device has to be placed on different parts of the body sequentially. The ECG signals measured from different leads are asynchronous. Artificial intelligence (AI) models for asynchronous ECG signals have barely been explored. OBJECTIVE: We aimed to develop an AI model for detecting acute myocardial infarction using asynchronous ECGs and compare its performance with that of the automatic ECG interpretations provided by a commercial ECG analysis software. We sought to evaluate the feasibility of implementing multiple lead-based AI-enabled ECG algorithms on smartwatches. Moreover, we aimed to determine the optimal number of leads for sufficient diagnostic power. METHODS: We extracted ECGs recorded within 24 hours from each visit to the emergency room of Ajou University Medical Center between June 1994 and January 2018 from patients aged 20 years or older. The ECGs were labeled on the basis of whether a diagnostic code corresponding to acute myocardial infarction was entered. We derived asynchronous ECG lead sets from standard 12-lead ECG reports and simulated a situation similar to the sequential recording of ECG leads via smartwatches. We constructed an AI model based on residual networks and self-attention mechanisms by randomly masking each lead channel during the training phase and then testing the model using various targeting lead sets with the remaining lead channels masked. RESULTS: The performance of lead sets with 3 or more leads compared favorably with that of the automatic ECG interpretations provided by a commercial ECG analysis software, with 8.1%-13.9% gain in sensitivity when the specificity was matched. Our results indicate that multiple lead-based AI-enabled ECG algorithms can be implemented on smartwatches. Model performance generally increased as the number of leads increased (12-lead sets: area under the receiver operating characteristic curve [AUROC] 0.880; 4-lead sets: AUROC 0.858, SD 0.008; 3-lead sets: AUROC 0.845, SD 0.011; 2-lead sets: AUROC 0.813, SD 0.018; single-lead sets: AUROC 0.768, SD 0.001). Considering the short amount of time needed to measure additional leads, measuring at least 3 leads-ideally more than 4 leads-is necessary for minimizing the risk of failing to detect acute myocardial infarction occurring in a certain spatial location or direction. CONCLUSIONS: By developing an AI model for detecting acute myocardial infarction with asynchronous ECG lead sets, we demonstrated the feasibility of multiple lead-based AI-enabled ECG algorithms on smartwatches for automated diagnosis of cardiac disorders. We also demonstrated the necessity of measuring at least 3 leads for accurate detection. Our results can be used as reference for the development of other AI models using sequentially measured asynchronous ECG leads via smartwatches for detecting various cardiac disorders.

Assessment of General Toxicity of the Glycyrrhiza New Variety Extract in Rats.

The Glycyrrhiza radix (Licorice) is one of the most commonly used medicinal plants in Asian countries, such as China, India, and Korea. It has been traditionally used to treat many diseases, including cough, cold, asthma, fatigue, gastritis, and respiratory tract infections. A Glycyrrhiza new variety, Wongam (WG), has been developed by the Korea Rural Development Administration and revealed pharmacological effects. However, the potential adverse effects of WG have not been revealed yet. This study evaluates the general toxicity of the WG extract through a single and repeated oral dose toxicity study in Sprague-Dawley rats. After single oral dose administration, no significant toxicological changes or mortality was observed up to 5000 mg/kg. Over a 4-week repeated oral dose toxicity study, no adverse effects and target organs were observed up to 5000 mg/kg/day. Over a 13-week repeated oral dose toxicity study, no mortality or toxicological changes involving ophthalmology, water consumption, or hematology were observed up to 5000 mg/kg/day. Although other parameters were changed, the alterations in question were not considered toxicologically significant, since responses remained within normal ranges and were not dose-dependent. In conclusion, the no-observed-adverse-effect level (NOAEL) of WG was higher than 5000 mg/kg/day, and no target organs were identified in rats.

Inhibitory Effect of Quercetin on Propionibacterium acnes-induced Skin Inflammation.

Quercetin is a well-known antioxidant and a plant polyphenolic of flavonoid group found in many fruits, leaves, and vegetables. Propionibacterium acnes is a key skin pathogen involved in the progression of acne inflammation. Although quercetin has been applied to treat various inflammatory diseases, the effects of quercetin on P. acnes-induced skin inflammation have not been explored. This study investigated the effects of quercetin on P. acnes-induced inflammatory skin disease in vitro and in vivo. The results showed that quercetin suppressed the production of pro-inflammatory cytokines in P. acnes-stimulated HaCaT, THP-1 and RAW 264.7 cells. Additionally, quercetin reduced the production of TLR-2 and the phosphorylation of p38, ERK and JNK MAPKs in P. acnes-stimulated HaCaT and THP-1 cells. It also suppressed MMP-9 mRNA levels in two cell lines exposed to P. acnes in vitro. In the case of in vivo, P. acnes was intradermally injected into the ears of mice and it resulted in cutaneous erythema, swelling, and a granulomatous response. Treatment with quercetin markedly reduced ear thickness and swelling. These results suggested that quercetin can be a potential therapeutic agent against P. acnes-induced skin inflammation and may have diverse pharmaceutical and cosmetics applications.

Production of protein hydrolysate from Protaetia brevitarsis seulensis (Kolbe) larvae by enzyme treatment under high pressure.

To improve the industrial use of health-functional materials based on edible insects, the objective of this study was to establish optimal conditions for improving the quality of Protaetia brevitarsis seulensis larval (PBSL) hydrolysates. PBSL was extracted using four methodologies: atmospheric pressure 50 °C-water extraction, atmospheric pressure 95 °C-water extraction, atmospheric pressure 50 °C-water enzymatic hydrolysis, and enzyme treatment under high pressure (HPE). The quality characteristics of soluble solid content, extraction yield, total protein content, protein yield, protein content with low molecular weight (LMW) (< 1kD), and the amino acid composition of hydrolysates were compared based on the different methods. All of the quality characteristics were found to be higher for HPE extracts than for the other extracts. Under optimized HPE conditions, extraction yield, protein yield, protein content with LMW, amino acid content and the content of essential amino acids increased by 3.4, 4.4 1.4 1.5, and 1.3 times respectively, compared to the other methods.

Comparison between covert sound-production task (sound-imagery) vs. motor-imagery for onset detection in real-life online self-paced BCIs.

BACKGROUND: Even though the BCI field has quickly grown in the last few years, it is still mainly investigated as a research area. Increased practicality and usability are required to move BCIs to the real-world. Self-paced (SP) systems would reduce the problem but there is still the big challenge of what is known as the 'onset detection problem'. METHODS: Our previous studies showed how a new sound-imagery (SI) task, high-tone covert sound production, is very effective for onset detection scenarios and we expect there are several advantages over most common asynchronous approaches used thus far, i.e., motor-imagery (MI): 1) Intuitiveness; 2) benefits to people with motor disabilities and, especially, those with lesions on cortical motor areas; and 3) no significant overlap with other common, spontaneous cognitive states, making it easier to use in daily-life situations. The approach was compared with MI tasks in online real-life scenarios, i.e., during activities such as watching videos and reading text. In our scenario, when a new message prompt from a messenger program appeared on the screen, participants watching a video (or reading text, browsing images) were asked to open the message by executing the SI or MI tasks, respectively, for each experimental condition. RESULTS: The results showed the SI task performed statistically significantly better than the MI approach: 84.04% (SI) vs 66.79 (MI) True-False positive rate for the sliding image scenario, 80.84% vs 61.07% for watching video. The classification performance difference between SI and MI was found not to be significant in the text-reading scenario. Furthermore, the onset response speed showed SI (4.08 s) being significantly faster than MI (5.46 s). In terms of basic usability, 75% of subjects found SI easier to use. CONCLUSIONS: Our novel SI task outperforms typical MI for SP onset detection BCIs, therefore it would be more easily used in daily-life situations. This could be a significant step forward for the BCI field which has so far been mainly restricted to research-oriented indoor laboratory settings.

Xanthone suppresses allergic contact dermatitis in vitro and in vivo.

Xanthone is a phenolic compound found in a few higher plant families; it has a variety of biological activities, including antioxidant, anti-inflammatory, and anticancer properties. However, the molecular and cellular mechanisms underlying the activity of xanthone in allergic contact dermatitis (ACD) remain to be explored. Therefore, this study aimed to investigate the regulatory effects of xanthone in ACD in human keratinocytes (HaCaT cell), and human mast cell line (HMC-1 cell) in vitro and in an experimental murine model. The results demonstrated that treatment with xanthone reduced the production of pro-inflammatory cytokines and chemokines including interleukin (IL)-1ß, IL-6, IL-8, and expression of chemokines thymus and activation-regulated chemokine (TARC) and macrophage-derived chemokine (MDC) in tumor necrosis factor (TNF)-α and interferon (IFN)-γ-stimulated HaCaT cells. Xanthone also suppressed the production of pro-inflammatory cytokines, chemokines, and allergic mediators in phorbol myristate acetate/A23187 calcium ionophore (PMACI)-stimulated HMC-1 cells. Xanthone significantly suppressed the phosphorylation of mitogen-activated protein kinases (MAPKs) and nuclear factor-kappa B (NF-κB) and activation of caspase-1 signaling pathway in vitro model. Additionally, xanthone administration alleviated 2,4-dinitrofluorobenzene (DNFB)-induced atopic dermatitis like-skin lesion by reducing the serum levels of immunoglobulin E (IgE), histamine, and pro-inflammatory cytokines and suppressing MAPKs phosphorylation. Xanthone administration also inhibited mortality due to compound 48/80-induced anaphylactic shock and suppressed the passive cutaneous anaphylaxis (PCA) reaction mediated by IgE. Collectively, these results suggest that xanthone has a potential for use in the treatment of allergic inflammatory diseases.

Cyanidin 3-Rutinoside, an Anthocyanin Pigment of Schisandra chinensis Baill, Inhibits Allergic Inflammation.

The occurrence of allergy-mediated inflammatory diseases such as asthma and atopic dermatitis have increased, but comprehensive treatment remains difficult. Previous studies have shown that Schisandra chinensis Baill has antioxidant, antidiabetic, and antitumorigenic effects. Cyanidin 3-rutinoside (CR) is the major anthocyanin pigment of S. chinensis. However, the biological effects of CR have been rarely studied to date. Therefore, the aim of this study was to investigate the regulatory effects of CR on phorbol-12-myristate-13-acetate (PMA)/A23187-induced allergic inflammation in vitro. CR inhibited the secretion of inflammatory cytokines such as interleukin-6 and tumor necrosis factor-α, and it also suppressed the phosphorylation of nuclear factor-kappa B. These results show that CR ameliorated PMA/A23187-induced allergic inflammation via the suppression of inflammatory cytokines in HMC-1 cells. Therefore, CR has potential as a therapeutic agent for allergic diseases.

A Novel Technique for Selecting EMG-Contaminated EEG Channels in Self-Paced Brain-Computer Interface Task Onset.

Electromyography artifacts are a well-known problem in electroencephalography studies [brain-computer interfaces (BCIs), brain mapping, and clinical areas]. Blind source separation (BSS) techniques are commonly used to handle artifacts. However, these may remove not only the EMG artifacts but also some useful electroencephalography (EEG) sources. To reduce this useful information loss, we propose a new technique for statistically selecting EEG channels that are contaminated with class-dependent EMG (henceforth called EMG-CCh). The EMG-CCh is selected based on the correlation between EEG and facial EMG channels. They were compared (using a Wilcoxon test) to determine whether the artifacts played a significant role in class separation. To ensure that the promising results are not due to the weak EMG removal, reliability tests were done In our data set, the comparison results between BSS artifact removal applied in two ways, to all channels and only to EMG-CCh showed that ICA, PCA, and BSS-CCA can yield significantly better ( ) class separation with the proposed method (79% of the cases for ICA, 53% for PCA, and 11% for BSS-CCA). With BCI competition data, we saw improvement in 60% of the cases for ICA and BSS-CCA. The simple method proposed in this paper showed improvement in class separation with both our data and the BCI competition data. There are no existing methods for removing EMG artifacts based on the correlation between the EEG and EMG channels. Also, the EMG-CCh selection can be used on its own or it can be combined with pre-existing artifact handling methods. For these reasons, we believe that this method can be useful for other EEG studies.

A novel onset detection technique for brain-computer interfaces using sound-production related cognitive tasks in simulated-online system.

OBJECTIVE: Self-paced EEG-based BCIs (SP-BCIs) have traditionally been avoided due to two sources of uncertainty: (1) precisely when an intentional command is sent by the brain, i.e., the command onset detection problem, and (2) how different the intentional command is when compared to non-specific (or idle) states. Performance evaluation is also a problem and there are no suitable standard metrics available. In this paper we attempted to tackle these issues. APPROACH: Self-paced covert sound-production cognitive tasks (i.e., high pitch and siren-like sounds) were used to distinguish between intentional commands (IC) and idle states. The IC states were chosen for their ease of execution and negligible overlap with common cognitive states. Band power and a digital wavelet transform were used for feature extraction, and the Davies-Bouldin index was used for feature selection. Classification was performed using linear discriminant analysis. MAIN RESULTS: Performance was evaluated under offline and simulated-online conditions. For the latter, a performance score called true-false-positive (TFP) rate, ranging from 0 (poor) to 100 (perfect), was created to take into account both classification performance and onset timing errors. Averaging the results from the best performing IC task for all seven participants, an 77.7% true-positive (TP) rate was achieved in offline testing. For simulated-online analysis the best IC average TFP score was 76.67% (87.61% TP rate, 4.05% false-positive rate). SIGNIFICANCE: Results were promising when compared to previous IC onset detection studies using motor imagery, in which best TP rates were reported as 72.0% and 79.7%, and which, crucially, did not take timing errors into account. Moreover, based on our literature review, there is no previous covert sound-production onset detection system for spBCIs. Results showed that the proposed onset detection technique and TFP performance metric have good potential for use in SP-BCIs.

Genetic Differences of Three Pollicipes mitella Populations Identified by PCR Analysis.

Genomic DNAs were extracted from the turtle leg (Pollicipes mitella, 1798) population of Tongyeong, Yeosu and Manjaedo located in the southern sea of Korea. The turtle leg population from Tongyeong (0.929) exhibited higher bandsharing values than did turtle leg from Manjaedo (0.852). The higher fragment sizes (>1,200 bp) are much more observed in the Yeosu population. The number of unique loci to each population and number of shared loci by the three populations, generated by PCR using 7 primers in the turtle leg (P. mitella) population of Tongyeong, Yeosu and Manjaedo. Genetic distances among different individuals of the Tongyeong population of the turtle leg (lane 1-07), Yeosu population of the turtle leg (lane 08-14) and Manjaedo population of the turtle leg (lane 15-21), respectively, were generated using the CLASSIFICATION option in Systat version 10 according to the bandsharing values and similarity matrix. The dendrogram, obtained by the seven decamer primers, indicated three genetic clusters: cluster 1 (TONGYEONG 01-TONGYEONG 07), cluster 2 (YEOSU 08-YEOSU 14), and cluster 3 (MANJEDO 15-MANJEDO 21). Tongyeong population could be evidently discriminated with the other two Yeosu and Manjaedo populations among three populations. The longest genetic distance (0.305) was found to exist between individuals' no. 02 of the Tongyeong population and no. 13 of the Yeosu population. It seems to the authors that this is a result of a high degree of inbreeding in narrow region for a long while.

Achieving network level privacy in Wireless Sensor Networks.

Full network level privacy has often been categorized into four sub-categories: Identity, Route, Location and Data privacy. Achieving full network level privacy is a critical and challenging problem due to the constraints imposed by the sensor nodes (e.g., energy, memory and computation power), sensor networks (e.g., mobility and topology) and QoS issues (e.g., packet reach-ability and timeliness). In this paper, we proposed two new identity, route and location privacy algorithms and data privacy mechanism that addresses this problem. The proposed solutions provide additional trustworthiness and reliability at modest cost of memory and energy. Also, we proved that our proposed solutions provide protection against various privacy disclosure attacks, such as eavesdropping and hop-by-hop trace back attacks.

Efficient management design for swimming exercise treatment.

Exercise-mediated physical treatment has attracted much recent interest. In particular, swimming is a representative exercise treatment method recommended for patients experiencing muscular and cardiovascular diseases. The present study sought to design a swimming-based exercise treatment management system. A survey questionnaire was completed by participants to assess the prevalence of muscular and cardiovascular diseases among adult males and females participating in swimming programs at sport centers in metropolitan regions of country. Using the Fuzzy Analytic Hierarchy Process (AHP) technique, weighted values of indices were determined, to maximize participant clarity. A patient management system model was devised using information technology. The favorable results are evidence of the validity of this approach. Additionally, the swimming-based exercise management system can be supplemented together with analyses of weighted values considering connectivity between established indices.

Intrusion-aware alert validation algorithm for cooperative distributed intrusion detection schemes of wireless sensor networks.

Existing anomaly and intrusion detection schemes of wireless sensor networks have mainly focused on the detection of intrusions. Once the intrusion is detected, an alerts or claims will be generated. However, any unidentified malicious nodes in the network could send faulty anomaly and intrusion claims about the legitimate nodes to the other nodes. Verifying the validity of such claims is a critical and challenging issue that is not considered in the existing cooperative-based distributed anomaly and intrusion detection schemes of wireless sensor networks. In this paper, we propose a validation algorithm that addresses this problem. This algorithm utilizes the concept of intrusion-aware reliability that helps to provide adequate reliability at a modest communication cost. In this paper, we also provide a security resiliency analysis of the proposed intrusion-aware alert validation algorithm.