中大「多種調控機制」 研基因

中大研究團隊花約一年半時間，將機器學習和自然語言處理等人工智能技術應用於基因表達調控的研究，開發嶄新的「嵌入式基因表現框架」（Gene Expression Embedding frameworK，簡稱GEEK），可同時研究多種調控機制對基因表達的影響，突破以往只考慮單一或小量機制的傳統研究模式。

Date: 
Monday, August 31, 2020
Media: 
HKET Daily
Media LInk: 
https://is.gd/bBABN7

New AI Approach to Investigate Multiple Gene Regulatory Mechanisms Concurrently For the Advancement of Biomedical Research

Date: 
2020-08-31
Thumbnail: 
Body: 
 
A research team from the Department of Computer Science and Engineering has developed a new Gene Expression Embedding frameworK (GEEK), which uses artificial intelligence technologies in machine learning and natural language processing to study the regulation of gene expression. In contrast to previous works that focused on one or a few regulatory mechanisms at a time, this new framework can study the joint effects of many mechanisms simultaneously. A research article describing this new study has been published in the renowned international science journal Nature Machine Intelligence. The framework may help study the causes of cancers and treatment methods.
 
Each human body contains tens of trillions of cells. While they mostly share the same DNA sequences, their gene activities can be markedly different. Such activities, referred to as “gene expression”, are affected by many regulatory mechanisms, such as transcription factor binding and protein interactions. In 2017, Prof. Kevin Yip from CUHK CSE and his research team studied one of the mechanisms that involves regulatory elements called enhancers. They investigated how enhancers are related to gene expression, and applied the results to discover three genes potentially related to liver cancer. This and other similar studies considered only individual gene regulatory mechanisms, and therefore could not fully understand the complex interplay between different mechanisms.
 
Prof. Yip used a metaphor to explain the intricate relationships among gene regulatory mechanisms. He said, “If you fail to turn on an electronic appliance using a remote controller, it seems like there is a problem with the controller, but the problem may also lie with the receiver or compatibility issues between the two. If we have a tool that can analyse the different components at the same time, it would be much easier to identify the root cause of the problem.”
 
The GEEK framework proposed by Prof. Yip's team makes use of machine learning and natural language processing methods, treating genes as “words” to capture their relationships in “sentences”. In the published study, GEEK was used to study several diverse gene regulatory mechanisms, including contacts in three-dimensional genome architecture, protein interactions, genomic neighborhoods and broad chromatin accessibility domains. The results showed that gene expression could be better explained when these mechanisms were modeled together than when they were considered separately.
Cancer is caused by mutations that lead to abnormal cell proliferation. “GEEK represents a novel way to study gene expression in different types of cells, including cancer cells,” says Prof Yip. “We will work closely with medical experts to try explaining some causes of liver cancer using GEEK. In the long run, we hope to extend our research to other cancer types and contribute to the development of new prevention and treatment methods.”
 
Among cancer treatments, immunotherapies are receiving a lot of attention due to their much greater efficacy in some cancer types. Yet the treatment outcome varies from patient to patient. Prof. Yip hopes that artificial intelligence can be used in the future to predict patients' responses to immunotherapies, which would improve treatment precision and reduce the burden on patients.
The research project was supported by the General Research Fund of the University Grants Council. Prof. Yip's team took one and a half years to produce the results. In the area of gene regulation research, Prof. Yip has more than ten years of experience, and he was one of the first to use machine learning and natural language processing to study gene regulation.

Quisque rutrum. Aenean imperdiet. Etiam ultricies nisi vel augue.

Quisque rutrum. Aenean imperdiet. Etiam ultricies nisi vel augue.

 

 

Filter: Dept: 
Faculty
CSE
Media Release

Physics Placement Test

Purpose of the test 
 
New entrants admitted without HKDSE results (regardless of admission channel and programme of admission) are required to sit for the placement test. Based on the results of the placement test, students will be assigned to either PHYS1003 or PHYS1110.   
 
Important Notes
Students attending the placement test will be pre-assigned to PHYS1110F as an interim arrangement.
 
Test format
1. There will be five long questions subdivided into shorter multiple-choice questions and numerical questions. The time allowed is 60 minutes. 
2. Please bring a HKEAA approved calculator.
3. This will be a closed-book test.
 
Instructions
Prior preparation
1. Students should prepare their own answer sheets, A4 size without any background design (other than faint lines if ruled paper is used). 
2. Please write full name (as it appears in CUSIS) and Student ID (if known) at the top of the first page, ahead of time. 
3. Log in ZOOM 15 minutes before the start of the placement test.  
 
ZOOM log-in identity
1. Your login name to ZOOM should be the same as your full name as it appears in CUSIS (e.g. John SMITH).  
 
During the examination
1. You should write your answers on paper first. You will be asked to copy the answers to Microsoft Forms at the end of the test.  
2. There are only two types of questions in the test: multiple choice questions and numeric questions. 
    (i) For multiple choice question, select the correct answer. 
    (ii) For numeric questions, some examples are shown below. 
       
        (a) Note that the unit is given in the question. You only need to input the numerical answers (Important: the system does not accept any fractions. Your final answer must consist of numerical values only)
        Example:
        The length of each side of a square is 2 m. What is the circumference of the square in metres? 
        You should type in 8 in the answer. 
       
        (b) Note that you should give your answer to 3 significant figures.
        Example:
        If your final answer is 3.414334….., please enter 3.41 in your answer box.
        (In this example, we will set a tolerance of +/- 0.1. The system accepts any value between 3.31 and 3.51 as a correct answer in this question.) 
 
        (c) You must agree with the honour pledge below before starting the online placement test. 
        “I affirm that I will not give or receive any unauthorized help on this exam, and that all work will be my own.” 
        Please choose “I agree” in the drop-down list before starting the test. 
 
Finishing the test 
1. Invigilators will announce the end of the test. You must stop writing immediately.  
2. After the end of the placement test, you will be given additional 5 minutes to transfer your answer to Microsoft Forms.  
3. You must not leave the ZOOM meeting room until you are told to do so. 
4. Please keep your answer scripts in a safe place for a week. You may be asked to send your answer script to the markers for verification. 
 
Announcement of placement test results and physics class assignment  
1. Based on the results of the placement test, you will be assigned to a physics course. A notification email will be sent to your CUHK 365 email account by 8th September 2020. It may take some time to update your course enrolment information in CUSIS. Please attend the assigned class stated in the email starting from 9th September 2020 (Tue).   
 
Sample questions
 
(a) Calculate the velocity of the two blocks immediately after the collision (in m/s) (10 marks)
(b) Calculate the maximum angle θ (in degrees) reached by the two blocks after the collision (10 marks)
 
Please enter the value “2.24” in Microsoft Forms.
Please enter the value “60.7” in Microsoft Forms. 
Venue
Online Test
Date: 
Thursday, September 3, 2020
Time
Thursday, September 3, 2020 to 15:00
e_title: 
Physics Placement Test
Not Available
Allow Regsiter: 

「深度偽造」流行 影片假可亂真

YouTube上載了一套「麻省理工」（MIT）入門「深度學習」（Deep Learning, DL）課程6.S191的視頻，其2020年版本的觀賞率在短短半年時間之內高達55萬次之多。課程第一堂，MIT主講老師阿米尼（Alexander Amuni）教授「邀請」了一名重要人物（VIP）作開場介紹，他便是前美國總統奧巴馬（Barack Obama）。雖然總統在屏幕前只是出現了片刻，但他竟能滔滔不絕、繪聲繪色地介紹深度學習的優點。

Date: 
Tuesday, August 25, 2020
Media: 
HKEJ
Media LInk: 
https://is.gd/MG95Sq

Mathematics Placement Test

Purpose of the Test

The Mathematics Placement Test is a pass-fail test! It tests a student's mathematical ability for studying calculus and other areas of mathematics in engineering. The test results are used to determine eligibility for enrollment in MATH1510 (Calculus for Engineers) in order to complete one of the graduation requirements for an Engineering programme.

Students who fail the test must take both MATH1020 (General Mathematics) and MATH1510 (Calculus for Engineers) in Term 1, 2020-21. Students who pass the test will only have to take MATH1510.

There are no exemptions made for the placement test!

Placement Test Schedule

The placement test is scheduled for Thursday, 20th August 2020. It will be held online via WeBWork. Please ensure that you have a stable internet connection. Students will be notified of the link to the test, login information and the exact test time individually by email.

 

General Characteristics of the Test

  1. The test consists two parts. In each part, there are 15 multiple-choice questions.
  2. Notes and calculators are permitted during the test.
  3. A student who for medical or other compelling reasons is unable to sit for the placement test shall apply with documentary evidence by email to math1510@math.cuhk.edu.hk on or before 21st August 2020 (Friday). No late request or request without supporting documents will be considered. Family obligations, summer jobs, job interviews or other unexpected events are not acceptable reasons in general and will result in a fail grade for the placement test.
  4. If a student's application for absence is approved, an online make-up placement test, together with an online oral exam will be given. The details are to be announced. 

 

Test Description

Below is a list of topics covered in the Mathematics Placement Test, which can be used as a study guide.

  1. Functions (including but not limited to:)
  1. Domain
  2. Range
  3. Composite
  4. Inverse
  5. Piecewise
  6. Graphs

 

  1. Polynomials and Rational Functions (including but not limited to:)
  1. Factorial
  2. Binomial Coefficients
  3. Binomial Theorem
  4. Partial Fraction

 

  1. Exponential and Logarithmic Functions (including but not limited to:)
  1. Euler's Number

 

  1. Trigonometry (including but not limited to:)
  1. Radian
  2. Inverse Trigonometric Functions

 

  1. Sequence, Series and Power Series (including but not limited to:)
  1. Summation Sign
  2. Radius of Convergence

 

  1. Vectors (including but not limited to:)
  1. Vectors in 2D, 3D
  2. Dot, Cross Product

 

  1. Probability (including but not limited to:)
  1. Mean, Variance, Standard Deviation
  2. Binomial
  3. Geometric
  4. Poisson Distribution

 

  1. Comples Numbers (including but not limited to:)
  1. Polar Form
  2. Argand Diagram
  3. De Moivre's Theorem

 

  1. Miscellaneous (including but not limited to:)
  1. Algebric Simplification
  2. Law of Indices
  3. Isolating Term
  4. Quadratic Formula
  5. Completing Square
  6. Absolute Value
  7. Rationalization
  8. Difference Quotient
  9. Inequality

 

Venue
Details will be sent through email
Date: 
Thursday, August 20, 2020
Time
Thursday, August 20, 2020
e_title: 
Mathematics Placement Test
Not Available
Allow Regsiter: 

Academic Counselling Session for New Students

Welcome to CUHK Engineering.

To provide students with a better understanding of the overall study arrangements of Engineering programmes, new students admitted to the Engineering Faculty should attend the Academic Counselling Session.

Organiser: 
Hosted by: Faculty of Engineering
Venue
ZOOM (Information will be provided by email)
Date: 
Tuesday, August 25, 2020
Time
Tuesday, August 25, 2020 to 17:45
e_title: 
Academic Counselling Session
Not Available
Allow Regsiter: 
Class: 
Current Student

Lu Yi-Chun opens up a new generation of batteries

Lithium-ion batteries are used to power everything from the mobile phone in your pocket to a Tesla truck. They pack a powerful current punch, making them very versatile for devices that need a brief, heavy jolt of electricity, like a digital camera.

Date: 
Monday, July 27, 2020
Media: 
Asia Today
Media LInk: 
http://www.asiatoday.com/pressrelease/lu-yi-chun-opens-new-generation-batteries

A Breakthrough by CUHK Engineering and University of Warwick Teams T-ray Camera Speed Boosted a Hundred Times Over

Date: 
2020-07-23
Thumbnail: 
Body: 

A research team from The Chinese University of Hong Kong (CUHK) and the University of Warwick has reached a crucial milestone towards developing single-pixel terahertz radiation (T-ray) imaging technology. Their single-pixel T-ray camera reached 100 times faster acquisition than the previous state-of-the-art without adding any significant costs to the entire system or sacrificing the sub-picosecond temporal resolution needed for the most sought-after applications, potentially opening the opportunity for them to be used in non-invasive security and medical screening. The breakthrough has been published in the journal Nature Communications.

The potentials and problems of Terahertz radiation

Terahertz (THz) radiation, or T-rays, sit in-between infrared and Wi-Fi on the electromagnetic spectrum. T-rays have different properties from other electromagnetic waves, most notably they can see through many common materials such as plastics, ceramics and clothes, making them potentially useful in non-invasive inspections. Another quality is that the low-energy photons of T-rays are non-ionizing, making them very safe in biological settings including security and medical screening. They are also highly sensitive to water and can observe minute changes to the hydration state of biological matter. This means that diseases perturbing the water content of biological matter, such as skin cancer, can potentially be detected using T-rays in vivo without any histological markers.

Efficient detection and generation of T-rays has been possible in laboratory settings for the last 25 years. However, THz technology is still not widely used in commercial settings as the cost, robustness and/or ease of use is still lagging behind for commercial adoption in industrial settings.

The advantages of single-pixel cameras

Professor Emma Pickwell-Macpherson from the Department of Electronic Engineering at CUHK and the Department of Physics at University of Warwick, said: “We use what is called ‘a single-pixel camera’ to obtain our images. In short, we spatially modulate the THz beam and shine this light onto an object. Then, using a single-element detector, we record the light that is transmitted (or reflected) through the object we want to image. We keep doing this for many different spatial patterns until we can mathematically reconstruct an image of our object.”

The researchers have to keep changing the shape of the THz beam many times which means this method is usually slower compared to multi-pixel detector arrays. However, multi-pixel arrays for the terahertz regime usually lack sub-picosecond temporal resolution, require cryogenic temperatures to operate or incur large equipment costs (>US$ 350,000). The setup developed by the CUHK Engineering and Warwick team, which is based on a single-element detector, is reasonably priced (~US$20,000), robust, has sub-picosecond temporal resolution (needed for accurate diagnosis) and operates at room temperature.

Professor Pickwell-Macpherson adds: “Our latest work improves upon the acquisition rate of single-pixel terahertz cameras by a factor of 100 from the previous state-of-the-art, acquiring a 32x32 video at 6 frames-per-second. We do this by firstly determining the optimal modulation geometry, secondly by modelizing the temporal response of our imaging system for improvement in signal-to-noise, and thirdly by reducing the total number of measurements with compressed sensing techniques. In fact, part of our work shows that we can reach a five times faster acquisition rate if we have sufficient signal-to-noise ratio.”

The research team led by Professor Pickwell-Macpherson has previously developed several THz devices including THz modulators that make use of the total internal reflection geometry to achieve high modulation depths across a broadband frequency range and a new approach for amplitude and phase modulation exploiting the Brewster angle. They are also working to improve the resolution of single pixel THz imaging through signal processing approaches. Future work will focus on improving the signal-to-noise and optimizing the software needed for accurate medical diagnosis, with the ultimate goal being to use single-pixel THz imaging for in vivo cancer diagnosis.

The research paper has been published in Nature Communications: https://doi.org/10.1038/s41467-020-16370-x

Read more about the research: http://bme.ee.cuhk.edu.hk/thzgroup/index.php and here: go.warwick.ac.uk/ultrafast

 

 

 

 

 

Professor Emma Pickwell-Macpherson.
.

 

Filter: Dept: 
Faculty
EE
Media Release

Can Speech AI Help Companies Save Cost?

Amidst the economic consequences caused by the global pandemic, automation has returned to the forefront of discussions. Across various industries, call centres and customer support are some of the sectors most vulnerable to automation, specifically to a branch of artificial intelligence (AI) called natural language processing — the art of making computers understand and communicate with human language.

Date: 
Wednesday, July 22, 2020
Media: 
LYNK
Media LInk: 
https://is.gd/3XL5FO

新冠疫情下的決策科學：群組測試

新冠肺炎威脅下，能否快速、大量地測試人們是否病毒攜帶者尤為重要。然而，香港至今仍未擁有足夠的測試能力。日前，香港大學盧寵茂教授指出香港需提升測試能力至每日二萬次。然而，現時實際測試水平為每日四千至五千次，與期望差距較大。我們或需安排更多資源（如測試儀器）以提升測試能力。不過，以決策科學的角度來看，我們可以嘗試更有效地運用現有資源，以提高測試能力。

Date: 
Wednesday, July 22, 2020
Media: 
Sing Tao Daily
Media LInk: 
https://is.gd/ehfnU6

Pages