top of page
yuhonghong resume_ページ_002.jpg
4080.jpg
c8b5fb6063b81383e9937c063ea1c7ca.jpg
yuhonghong resume_ページ_003.jpg
hocit999_ページ_4.jpg
yuhonghong resume_ページ_003.jpg

Welcome to the entrance to the world of acupuncture
Acupuncturist: a profession that will not disappear in the AI ​​era
With the development of AI (artificial intelligence), many existing occupations are expected to disappear. It is said that artificial intelligence has gradually entered the high-level intellectual professional fields such as finance and legal affairs, and has begun to serve people.

But what about acupuncture?

Acupuncture is a delicate and special technique that cannot be easily replaced by computers or machines. It can only be done manually. We often hear about their physical conditions from patients, consider the cause of the current symptoms, and select appropriate treatment points from 361 acupoints (acupoints) for acupuncture. This is a complex and delicate technique. It is impossible for robots. Even if it is possible, it will not be controlled by practitioners.

On the contrary, since it is surrounded by digital devices every day, do you think it needs the warmth, softness and safety of human hands? Acupuncture is a technology that is expected to become more valuable in the future AI era.

The WHO (World Health Organization) has also certified the therapeutic effects of acupuncture. Acupuncture therapy is good at treating symptoms that are difficult to improve by Western medicine, pain that has not been diagnosed as "discomfort" (discomfort) according to the examination results, and pain that cannot be completely cured. It is understandable that in recent years, the upsurge of Oriental medicine is coming, and it is said that this is a very stressful society. There is a growing demand for acupuncturists who can understand and treat patients' pain.

In addition, acupuncturists are long-term participants of men and women. This is not hard work, so even after aging, you can use the skills and intuition accumulated according to your physical strength to stand in the field of treatment. This is a lifelong job, you can improve your skills while you work, and cultivate opportunities for success while reaping rewards.
 

ACCELERATE OUTCOMES WITH A BACKBONE FOR ARTIFICIAL INTELLIGENCE.

makes machine learning operational by connecting models to the real-world decisions they inform. Often, AI/ML algorithms live in experimental vacuums. Foundry provides the end-to-end infrastructure an organization needs to apply AI/ML to real problems and real data:

  • A data foundation.  provides the data engineering capabilities an organization needs to deploy AI/ML models it can trust. Organizations use  to build a solid foundation of sufficient, quality data, then bring that data into daily operations.

  • Production deployment infrastructure. revolutionizes the way organizations build and deploy AI/ML by combining a data foundation with end-to-end algorithm deployment infrastructure. Data scientists and engineers can customize, deploy, assess, and compare across homegrown, open-source, and third-party algorithms. All models are tightly integrated with end-to-end platform capabilities, ranging from feature curation and health checks to model management to inference/serving to outcome monitoring.

  • Faster feedback loops. AI/ML models rarely work on a "set it and forget it" basis.  integrates the full model lifecycle with end user actions and feedback, and with operational decisions and outcomes. This enables operationally oriented model monitoring, management, understanding, selection, and adjustment. The result is more adaptable and ambitious AI/ML, faster.

UNITE THE ORGANIZATION AROUND A SHARED ENVIRONMENT FOR MACHINE LEARNING.

The most effective AI/ML encodes and supercharges an organization's unique expertise. That requires uniting an organization's data scientists, decision-makers, and everyday employees in an environment for collaborating on AI/ML-powered operations.  collaboration infrastructure drives AI/ML that brings the organization together:

  • A unifying ontology. ontology translates an organization's complex data landscape into human-readable concepts. built models reflect how an organization views the world and unite data scientists, engineers, analysts, executives, and operational end users around a common semantic layer.

  • Granular security controls. lets organizations define granular access control policies at the integration stage, then propagates those policies intelligently across the system. Organizations can promote collaboration confidently with granular data security and transparent data governance.

  • Model templates.model templates empower low-code and no-code AI/ML so even non-technical users can use AI to accelerate and enrich their workflows.

MAXIMIZE IMPACT AND MINIMIZE RISK WITH A PLATFORM DESIGNED FOR RESPONSIBLE ENGINEERING.

Our approach to AI/ML in  reflects our foundational belief in augmenting human intelligence, not replacing it. We believe AI/ML algorithms are most effective when they empower humans to ask complex questions, interpret answers, and act on results. At public and private sector organizations around the world, driven AI/ML is accelerating human decision-making by:

  • Surfacing new leads in dark web, weapons trafficking, financial fraud, and drug trafficking investigations so investigators can identify persons of interest more quickly

  • Aggregating and correlating biomedical research data to streamline drug discovery

  • Processing entity resolution suggestions so analysts can focus on making assessments rather than manually sorting and tagging data

  • Analyzing massive-scale sensor data so that engineers can make better aircraft maintenance decisions

  • Rapidly generating simulations while allowing operators to tweak scenario variables, leading to better-informed decisions optimized for different variables (e.g., safety rating and production quantity)

I see our approach to ethical machine learning as being grounded in an appreciation for both the promise and limitations of human-computer collaboration. The promise of AI is in augmenting and enhancing human intelligence, expertise and experience. Think helping a aircraft mechanic make better, more accurate and more timely repairs – not automating the mechanic out of the picture.

— Anthony Bak, Co-Lead of Palantir's Machine Learning team, in an interview with CTOVision

Protecting privacy and preserving civil liberties is fundamental to our mission.ships with technical capabilities that enable ethical engineering and ethical machine learning, including data protection features such as granular access controls, data provenance and lineage tracking, data retention and deletion management, and audit logging.  also enables industry-leading monitoring and validation for AI models. Its flexible, configurable tools let organizations evaluate model bias, in terms of both the data used to train the model and model outcomes.

We recognize that technology alone can't fully mitigate the risks of machine bias. Our dedicated Privacy and Civil Liberties team works with our engineers and our customers to approach building and deploying AI/ML thoughtfully.

Acupuncture significantly reduces AI-associated arthralgias

Publish date: December 8, 2017

By 

Roxanne Nelson

 

 

 

REPORTING FROM SABCS 2017

SAN ANTONIO – Acupuncture significantly reduced joint pain that was associated with the use of aromatase inhibitors (AIs) in women with early breast cancer, according to new findings reported at the San Antonio Breast Cancer Symposium.

The randomized, phase 3 SWOG S1200 clinical trial found that, compared with sham acupuncture and a control group receiving no therapy, women receiving acupuncture reported significantly lower scores on the Brief Pain Inventory–Short Form (BPI).

“We have shown consistently, with multiple measures assessing pain and stiffness, that true acupuncture generated better outcomes than either control group in a large multicenter trial,” said lead author Dawn L. Hershman, MD, leader of the Breast Cancer Program at the Herbert Irving Comprehensive Cancer Center at NewYork-Presbyterian/Columbia University Medical Center. “Acupuncture provides a nonpharmacologic option that can improve symptoms and possibly increase AI adherence and subsequent breast cancer outcomes.”

AIs can reduce both early breast cancer recurrence and mortality. Dr. Hershman noted that these agents are effective in the adjuvant setting and for prevention “but we know that it doesn’t work if you don’t take it. Noncompliance is a major problem among women taking hormonal therapy.”

Noncompliance is multifactorial and one of the main reasons women discontinue their therapy early is because of arthralgias or joint discomfort. “We were interested in a nonpharmacologic intervention, to assess whether or not we could control these symptoms.”

Dr. Hershman pointed out acupuncture provides a safe and effective alternative for patients reluctant to take a prescription medication that can result in other side effects. “Identification of nonopioid options for pain control is a public health priority,” she said.

Acupuncture is a popular nonpharmacologic modality and widely used for a number of indications. Several single-institution studies have suggested that it may be useful for controlling AI-associated arthralgias, while other studies have not demonstrated a benefit.

In this trial, the authors evaluated the efficacy of acupuncture, compared with sham acupuncture or waitlist control, in the treatment of AI associated arthralgia in a large population of patients. The study was conducted at 11 centers.

The cohort comprised 226 postmenopausal women diagnosed with early-stage, hormone receptor–positive breast cancer who were receiving treatment with AIs. The primary endpoint was the decline in joint pain as measured by BPI-SF at 6 weeks, and to assess the duration of the effect, the women were followed for an additional 12 weeks.

Within this group, 110 were randomized to true acupuncture; 59 to sham acupuncture, and 57 to waitlist control (no treatment). Patients receiving true or sham acupuncture had sessions three times a week for 6 weeks followed by one session per week for 6 more weeks. Pain status was reported at baseline, during treatment, and then afterwards, using a variety of measurement tools including the BPI-SF, which is a self-administered 14-item questionnaire that evaluates pain severity on a 0-10 scale, and the impact of pain on activities of daily living.

At 6 weeks, the true acupuncture treatment arm reported significantly lower BPI worst pain scores than those in the sham acupuncture and the waitlist control arms. The mean BPI worst pain for the true acupuncture arm was 0.92 points lower than the sham acupuncture arm (P = .01) and 0.96 points lower than the waitlist control arm (P = .01). The proportion of patients experiencing a large reduction in BPI worst pain (greater than 2) was significantly greater in the true acupuncture arm, compared with the other groups: 58% versus 33% percent and 31%, respectively. The differences continued to remain statistically significant at 24 weeks, even though the treatment only continued for 12 weeks.

Associated adverse effects were minimal with true and sham acupuncture and limited to grade 1 bruising.

The cost of the 12-week intervention was about $1,250 or $65-$75 a session. “We feel that there is now sufficient evidence to support insurance coverage of acupuncture of AI arthralgia.”

In a discussion of the paper, Dr. Anne Partridge, from the Dana Farber Cancer Center, noted that it is imperative to seek new ways to improve outcomes in breast cancer, and AIs are contributing to that. However, she echoed the concern that nonadherence to treatment is a “tremendous problem” and hampers the clinical effectiveness of AI therapy.

The rate of discontinuation during the first year of therapy is 20% within the first year and up to 40% of patients do not take them daily. Both early discontinuation and nonadherence contribute to mortality.

Based on these results from the largest randomized controlled trial looking at acupuncture in this setting, should physicians be recommending acupuncture to patients prescribed AI therapy?

“The short answer is, why not?” said Dr. Partridge, “And that we should be recommending it for some of our patients.”

However, there are a number of issues that need to be addressed, she added. The duration of treatment is not known, and the need for follow-up treatment or the frequency of it is not known. The generalizability of it is also unclear when looking at a larger population, and acupuncture is highly operator dependent.

“There are cost and access issues, and insurance right now offers very limited coverage,” she said.

Importantly, Dr. Partridge emphasized, “We know that it will help symptoms, but will it improve adherence to AI?”

It may improve adherence for some patients, but “side effects are only one factor,” she said. “Adherence behavior is complicated. We need to figure out how to optimize these therapies in our patients.”

This study was supported by the National Institutes of Health National Center for Complementary and Integrative Health and the Office of Research on Women’s Health, and grants from the NIH/National Cancer Institute Division of Cancer Prevention. Dr. Hershman declared no conflicts of interest. Dr. Partridge had no disclosures.

empty-wheelchair-parked-in-hospital-pict
digitwin.png

Illumina, Chinese Children's Hospital to Launch Newborn Sequencing Study

Apr 23, 2019

 

| 

staff reporter

 

NEW YORK (GenomeWeb) – Illumina and the Children's Hospital of Fudan University in China plan to launch a study of whole-genome sequencing in the hospital's neonatal intensive care unit to determine whether it can be used as a diagnostic for critically ill infants, Illumina said this week.

According to Illumina, the researchers plan to enroll 200 patients and compare the diagnostic rate of rapid WGS with genetic diagnostic methods such as microarray analysis and gene panel sequencing.

The researchers will also compare the time it takes to reach a diagnosis, impact on the patient's prognosis, and turnaround time.

Illumina will provide the sequencing reagents and the Children's Hospital of Fudan University will conduct the testing and data analysis and also be responsible for reporting results and providing genetic consultations with family members.

The hospital is also a sponsor of the Newborn Genome Project, which is creating a genome database for newborns in China in order to develop better methods for detecting genetic diseases in newborns and to establish standards for neonatal genetic diseases.

UK Study Details Liquid Biopsy's Ability to Guide Metastatic Breast Cancer Treatment

Dec 13, 2019

 

| 

staff reporter

 

NEW YORK – Researchers from the Institute of Cancer Research, London and the Royal Marsden NHS Foundation Trust have presented data demonstrating that blood-based genotyping assays can accurately detect specific biomarkers and potentially be used to guide targeted treatment of metastatic breast cancer. 

The researchers used Guardant Health's liquid biopsy assay and Bio-Rad Technologies' droplet digital PCR (ddPCR) testing to identify errors — including mutations in the HER2, ESR1 and AKT1 genes — in circulating tumor DNA (ctDNA) that had been shed into the patients' bloodstream. 

During a presentation this week at the San Antonio Breast Cancer Symposium (SABCS), Nicholas Turner, molecular oncology professor at the Institute of Cancer Research, explained that his team wanted to solve the issue of genotyping breast cancer tumors without having to perform multiple biopsies. Turner highlighted the need for prospective studies to assess the accuracy of ctDNA testing in routine practice and the potential of these tools to guide targeted therapy without requiring solid tissue testing. 

With funding from Stand Up To Cancer (a fundraising campaign from Cancer Research UK and Channel 4) Turner's team therefore launched an ongoing prospective study, called "plasmaMATCH," to examine the clinical utility of ctDNA for metastatic breast cancer detection. The group enrolled a total of 1,044 patients with advanced breast cancer who had either progressed on prior drug therapy or relapsed within a year after adjuvant chemotherapy. 

The researchers analyzed ctDNA in blood samples from patients using the Guardant360 sequencing-based assay and Bio-Rad's ddPCR as an orthogonal method. Patients with identified mutations could also enroll in a treatment arm of the study that matched their mutation. 

Turner and his team split the patient population into three major cohorts depending on specific genetic errors found in ctDNA: ESR1 mutations, HER2 mutations, and AKT-1 mutations in estrogen-receptor-positive breast cancer. The researchers also added a fourth cohort that had AKT-1 and PTEN mutations based on both ctDNA and tumor sequencing results, as well as a fifth group with triple-negative breast cancer without mutations. 

While Turner's team's initially aimed to measure the response rate of targeted therapies matched to mutations in ctDNA without tissue testing, the researchers also monitored the frequency of mutations, accuracy of testing, proportion of patients entering a treatment cohort, and the activity in clonality-dominant versus subclonal mutations. Turner's team discovered that 784 patients who had cfDNA testing done with both Guardant360 and ddPCR had "very strong agreement" of between 96 and 99 percent. Comparing the ctDNA to the patients' tumor samples to validate the plasmaMATCH findings, the researchers found that the liquid biopsy assay had an overall sensitivity of 93 percent. 

Guardant360 also identified several more targetable alterations than hotspot testing, including over 35 percent more PIK3CA mutations that can be targeted by US Food and Drug Administration-approved therapies. The assay also detected significantly more ESR1 mutations and found previously undetected microsatellite instability and ERBB2 mutations.   

The researchers provided 142 patients that had specific identified mutations experimental targeted therapies as part of the study, and they plan to test the treatments that showed initial promise in larger clinical trials. 

The researchers now believe that they can use blood testing to identify rare subtypes of breast cancer, in addition to potentially replacing more invasive methods of breast cancer detection. 

"We have now confirmed that blood tests can quickly give us a bigger picture of the mutations [that] are present within multiple tumors throughout the study," Turner said in a statement.  

"We show that circulating tumor DNA testing offers a simple, efficient, and fast method of tumor genotyping," Turner also noted during the presentation at SABCS. "The patients with mutations identified in their ctDNA have efficacy with matching target[ed] therapies." 

The researchers believe the blood-based assays are now reliable enough to be applied routinely by clinicians after receiving regulatory approval.

"As the number of treatment-relevant genomic alterations in metastatic breast cancer continues to grow, it is critical that oncologists have a simple and reliable way to comprehensively test for this information about their patients," Guardant Health Global Chief Medical Officer Rick Lanman said in a statement. "Guardant360 detected changes in the genome picture over time … and also identifies targetable but uncommon genomic biomarkers."

1c8b9b577cc41266ebab5e65ff8f3de8.jpg
090320_COVID_Test_0111-1024x682.jpg
05b50bc878af8f403a01848035fa3a86.jpg
7d54e3018152c4d0b389e55596039404.jpg
4deac0802d4054a8a85b3f3d256911aa.jpg
hocit999 - コピー_ページ_16.jpg
hocit999_ページ_6.jpg
bottom of page