STOCK MARKET UPDATE

Ticker

6/recent/ticker-posts

Importance of Artificial Intelligence & Biophotonic Techniques in Point of Care Diagnosis of HIV/AIDS

Abstract :

Bio photonics is a branch of science dealing with the interaction of light in biological substances such as tissues and cells
at scales ranging from microns to the nano-level. This  quality of biophotonics leads to understand hidden knowledge of
cell-cell  interaction,  cell-tissue  interaction  and  so-on.  It  needs  to  be  more  explore  in  HIV/AIDS  for  treatment  and
diagnosis. Artificial Intelligence based predictive models help to develop such chips which are cost effective, ready to use
and require less time for diagnosis. These photonics based
methods also need to develop therapeutics for HIV. 

 Keywords: Biophotonics; Artificial intelligence; Therapeutics; HIV
Introduction 


 

 

    Introduction 

     Advancement in  technology is  of need  today.  With  the
growing cutting edge technology, diagnosis and treatment
time is  reduced.  Bio photonic  technology is  one  of  them.
The discipline of bio photonics deals with the interaction
of  light,  or  electromagnetic  radiations  with  living
organisms and biologically active macromolecules such as
proteins(hemoglobin),nucleic  acids  (DNA and  RNA), and
metabolites (glucose and lactose). In both high (x ray) and
low  (radio  frequency  (RF)  energies  the  body  is  almost
transparent,  this  allows the  non-invasive imaging  of  the
internal structure of  organs and bones. This focused light
of  lasers  with  colors  can  be  used  for  a  wide  variety  of
unique  therapeutic  interventions  of  specific  regions  of
organs and tissues.

     HIV  (human  immunodeficiency  virus)  detection  in
biological  samples  is  critical.  Recently  an  optical
biosensor  is  patented  that  is  able  to  detect  the  virus  a
week after being infected, with a total test time of 4 hours
and  45  minutes  thereby  allowing  clinical  results  to  be
obtained  on  the  same  day  [1].  The  biosensor  combines
micromechanical  silicon  structures  with  gold
nanoparticles,  both  functionalize  with  p24-specific
antibodies.  The  gold  nanoparticles  have  optical
resonances  known  as  Plasmon’s,  which  are  capable  of
scattering  light  very  efficiently.  Micromechanical
structures  are  excellent  mechanical  sensors  capable  of
detecting  interactions  as  small as  intermolecular  forces.
The  combination  of  these  two  structures produces  both
mechanical and optical  signals that  amplify one another,
producing  remarkable  sensitivity  to  detect  the  p24
(protein marker for HIV) [2].

     Detecting  and  quantifying  biomarkers  and  viruses  in
biological  samples  have  broad  applications  in  early
disease  diagnosis  and  treatment  monitoring.  It has  been
demonstrated that a label-free optical sensing mechanism
using nanostructured photonic  crystals  (PC) can  capture
and  quantify  intact  viruses  (HIV-1)  from  biologically
relevant samples [3,4
Bio  photonics  and  Artificial  Intelligence
Techniques
     With  the  applications  of optics  and photonics,  various
methods  are  developed  by  spectrophotometer,
microscopy,  and  lasers  etc.  for  disease  diagnosis  in  not
only molecular level but also in tissue level. Some of them
are described below:

 Hyper spectral Imaging (HSI): It is also called imaging
spectrometer which involves various medical applications
specifically  in  image  guided  surgery  and  diagnosis  of
diseases.  It  has  been  assumed  that  the  scattering,
fluorescence and absorption properties of  tissues vary as
the  disease  progresses.  Therefore,  the  transmitted,
reflected,  and  fluorescent  light  from  tissue  has  been
captured  by  HIS  having  quantitative  diagnostic  data.
Which can be used by artificial intelligence based machine
learning  techniques  to interpret  HIV  disease  and  use  of
ART in particular time [5].
 
 Diffuse optical Imaging: This technique is  classified in
two groups:
a) Diffuse  optical  topography---If  the  tissue  optical
properties get modified after a period of time, there are
possibilities for  the  photon  to reach same  detector  by
altering  the  measurable  intensity.  A  2D  topographical
data set can  be  constructed by  measuring  the changes
between every set of source and detector. This helps to
study HIV at a tissue level [6].
b) Reconstructed  topography—the  absorption  changes
determined  by  high  resolution  images  leads  to
production  of  3D  reconstruction  of  the  image.  It  is
usually  done  via  the  detection  of  the  absorption
distribution, where the data measured is matched with
simulation  results  of  numerous  absorption
distributions  inside  the  3D  images.  Hence,  the
recording  of  the  signal  at  various  distances  between
source  and  detector  is  essential  and  measured  [7].
These  results  would  be  used  input  for  Artificial
Intelligence  based  machine  learning  techniques  for
disease  diagnosis  in  particular  stage  as  well  as
molecular level of HIV infection is well understood.

 Diffuse  optical  tomography:  This  techniques  needs
computed  tomography  in  reconstructing  the  3D  images
where  recording  of  sequence  of tissue  measurements  is
performed. The 3D image has to be acquired from various
angles  which  will  give  complete  information  of  HIV
affected cells/tissues [8].

 Lasers: Its  better  if Lasers  system  is used with  sensor
control.  Because  sensor  controlled  laser  systems  are  of
focus in the field of therapy. This was previously used  in
treating cancer. But  it can  be  used for  monitoring of  HIV
infection spread in body.

 Flow  Cytometer (FC): A  flow  cytometer  is  a  machine
driven instrument that can be used to examine single cell
properties i.e., allow only one cell to be analyzed at a time.
It  can be  used to  measure the  cell  granularity,  cell  size,
and  to  quantify  various  cell  components  that  include
newly synthesized DNA and the total DNA, the number of
specific  cell  surface  receptors,  gene  expression  as  the
amount  of  messenger  RNA  for  a  particular  gene,  and
amounts  of  transient  signaling  events  and  intracellular
protein in living cells. Hence FC is most adaptable to study
HIV features. It is used in diagnosis of HIV disease as well
as its progression [6].
 

 Fluorescent  Markers:  A  significant  feature  of  Bio

photonics involves the visualization and detection of cells
and  tissues.  Which  includes  injection  of  fluorescent
markers, into a living system, to follow dynamics of a cell
and  drug  delivery?  This  visualization  and  detection  of
fluorescent markers  can be  counted  and  effect  on  living
system  is  shared  by  AI  based  ML  techniques.  By  this
method  cell  reaction  and  drug  delivery  to  tissues  are
monitored  in  specific  time.  This  can  be  used  in
Fluorescence  lifetime  imaging  microscopy  (FLIM).  The
information  obtained  by  FILM  is  used  in  local
environment sensing, detection of molecular interactions,
detection  of  conformational  changes,  discrimination  of
multiple  labels  or  background  removal,  tissue
characterization  by  auto  fluorescence  and
characterization and quality control of  new  materials. All
these parameters based study are required for input data
of  machine  learning  techniques.  And  our  expected
outcome  would  able  to  predict  which  parameter is  best
for  study  cell  based  interactions  of  HIV.  Confocal
microscopy  and  multiphoton microscopy  are also  found
to  be  suitable  for  HIV  cells.  Because  in  multiphoton
microscopy,  instantaneous  absorption  of  two  incident
photons from  a  pulsed infrared laser source is  observed.
So  this  will helpful  in  how a  particular HIV  infected  cell

response [7].

  Targeted  molecular  imaging:  It  involves  analyzingmicron-level biological processes. It is used to analyze theshape and role of the molecular system by generating the

signals  incident  from  the  molecules.  Therefore  the
produced image  describes  the  3-D spatial  distribution  of
the  targeted  molecules  in  the  tissue,  specifies  the
diagnostic  data  at  the  molecular  level,  and  shows  the
functional cell  properties.  These  properties  are  essential
to  study  HIV  infected  cells  at  molecular  level  with  its
functionality opens new avenues of HIV [8
     Other  methods  like  optical trapping, second  harmonic
trapping, and cell transfection are used to explore in case
of HIV.

     Artificial intelligence makes  our  computer to solve the
complicated  problem  by  training  and  testing  the  data
given by us. Hence our intelligence and computer learning
intelligence works  together to  develop the  model  or  our
knowledge for better use.

     Machine  learning  which  evolved  from  pattern
recognition and  computational learning theory, is  able  to
construct algorithms that can learn and make predictions
with  data.  There  are  many  machines  learning  software
tools.  We  use  decision  tree  induction  algorithms  and
Naïve bayes  algorithm  of WEKA  software package  [6]  to
classify and compare a given HIV data set. CD4+count and
IL-10,p24,  IFN-  biomarkers  were  used  to  determine
diagnosis  and  screening  of  HIV/AIDS  [9,10].  There
interaction  and  progression  during  HIV  course  was
detected by above discussed bio photonics methods:
 
                               Decision  tree A  decision  tree  (or  tree  diagram)  is  a
decision support tool that uses a tree-like graph or model
of  decisions  and  their  possible  consequences,  including
chance  event  outcomes,  resource  costs,  and  utility.
Another use of decision trees is as a descriptive means for
calculating  conditional  probabilities.  In  data  mining  and
machine  learning,  a  decision  tree  is a  predictive  model;
that  is,  a  mapping  from  observations  about  an  item  to
conclusions  about  its  target  value.  More  descriptive
names  for  such  tree  models  are  classification  tree
(discrete  outcome)  or  regression  tree  (continuous
outcome).  In  these  tree  structures,  leaves  represent
classifications  and  branches  represent  conjunctions  of
features  that  lead  to  those  classifications.  The  machine
learning technique for inducing a decision tree from data
is  called  decision  tree  learning,  or (colloquially) decision
trees [11,12]. With the help of symptoms of HIV infection,
and above  mentioned  biomarkers,  a decision  tree model
is  developed  which  would  tell  whether  the  person  is
infected with HIV and is in which stage. Accordingly anti-
retroviral  therapy  will  started.  As  this  is  very  fast  and
accurate it can be used in POC diagnostics for HIV/AIDS.

 Naïve-bayes:  This  classifier  is  a  simple  probabilistic
classifier  based  on applying Bayes'  theorem  with  strong
(naive)  independence  assumptions.  A  more  descriptive
term  for  the  underlying  probability  model  would  be
"independent  feature  model".  In  simple  terms,  a  naive
Bayes  classifier  assumes  that  the  presence  (or  lack  of
presence) of a particular feature of a class is unrelated to
the  presence  (or  lack  of  presence) of  any  other  feature     
Depending on the precise nature of the probability model,
naive Bayes classifiers can  be trained very efficiently in a
supervised  learning  setting.  In  many  practical
applications,  parameter  estimation  for  naive  Bayes
models uses the method of maximum likelihood [13,14]. 

     Other machine learning algorithms are also used based
on  the  data  provided  by  bio  photonics  techniques  and
accordingly  model  development  and  validation  can  be
done for further investigation.
  Targeted  molecular  imaging:  It  involves  analyzing
micron-level biological processes. It is used to analyze the
shape and role of the molecular system by generating the
signals  incident  from  the  molecules.  Therefore  the
produced image  describes  the  3-D spatial  distribution  of
the  targeted  molecules  in  the  tissue,  specifies  the
diagnostic  data  at  the  molecular  level,  and  shows  the
functional cell  properties.  These  properties  are  essential
to  study  HIV  infected  cells  at  molecular  level  with  its
functionality opens new avenues of HIV [8
     Other  methods  like  optical trapping, second  harmonic
trapping, and cell transfection are used to explore in case
of HIV.


     Artificial intelligence makes  our  computer to solve the
complicated  problem  by  training  and  testing  the  data
given by us. Hence our intelligence and computer learning
intelligence works  together to  develop the  model  or  our
knowledge for better use.
 

     Machine  learning  which  evolved  from  pattern
recognition and  computational learning theory, is  able  to
construct algorithms that can learn and make predictions
with  data.  There  are  many  machines  learning  software
tools.  We  use  decision  tree  induction  algorithms  and
Naïve bayes  algorithm  of WEKA  software package  [6]  to
classify and compare a given HIV data set. CD4+count and
IL-10,p24,  IFN-  biomarkers  were  used  to  determine
diagnosis  and  screening  of  HIV/AIDS  [9,10].  There
interaction  and  progression  during  HIV  course  was
detected by above discussed bio photonics methods:

 Decision  tree:  A  decision  tree  (or  tree  diagram)  is  a
decision support tool that uses a tree-like graph or model
of  decisions  and  their  possible  consequences,  including
chance  event  outcomes,  resource  costs,  and  utility.
Another use of decision trees is as a descriptive means for
calculating  conditional  probabilities.  In  data  mining  and
machine  learning,  a  decision  tree  is a  predictive  model;
that  is,  a  mapping  from  observations  about  an  item  to
conclusions  about  its  target  value.  More  descriptive
names  for  such  tree  models  are  classification  tree
(discrete  outcome)  or  regression  tree  (continuous
outcome).  In  these  tree  structures,  leaves  represent
classifications  and  branches  represent  conjunctions  of
features  that  lead  to  those  classifications.  The  machine
learning technique for inducing a decision tree from data
is  called  decision  tree  learning,  or (colloquially) decision
trees [11,12]. With the help of symptoms of HIV infection,
and above  mentioned  biomarkers,  a decision  tree model
is  developed  which  would  tell  whether  the  person  is
infected with HIV and is in which stage. Accordingly anti-
retroviral  therapy  will  started.  As  this  is  very  fast  and
accurate it can be used in POC diagnostics for HIV/AIDS.
 

Naïve-bayes:  This  classifier  is  a  simple  probabilistic
classifier  based  on applying Bayes'  theorem  with  strong
(naive)  independence  assumptions.  A  more  descriptive
term  for  the  underlying  probability  model  would  be
"independent  feature  model".  In  simple  terms,  a  naive
Bayes  classifier  assumes  that  the  presence  (or  lack  of
presence) of a particular feature of a class is unrelated to
the  presence  (or  lack  of  presence) of  any  other  feature     
Depending on the precise nature of the probability model,
naive Bayes classifiers can  be trained very efficiently in a
supervised  learning  setting.  In  many  practical
applications,  parameter  estimation  for  naive  Bayes
models uses the method of maximum likelihood [13,14]. 

     Other machine learning algorithms are also used based
on  the  data  provided  by  bio  photonics  techniques  and
accordingly  model  development  and  validation  can  be
done for further investigation.
 specific  HIV-1  diagnostic  tools  [7,17,18].  If  the  kit  is
developed  which  not  only  shows  viral  load  also  gives
genotype information then it would be breakthrough and
innovative research in the field of HIV/AIDS therapeutics.
Here the issues of accuracy cost and time were important
for the purpose of the research based on POC of HIV/AIDS
being  a  serious  disease.  To  resolve  the  issues  Artificial
intelligence  based  machine  learning  techniques  are
required to overcome the problem.


"Concluding Remark and Future Prospects
     Bio  photonics  techniques  increase  the  chances  for
understanding the  HIV  disease  better  at molecular  level.
With  the  handshake  of  AI  techniques,  biomedical
scientists  will  be  able  to  provide  environment  friendly
techniques as  they are fast and economical, accurate and
reliable  to  use.  The  data  of  molecular  analysis  of  Bio
photonics techniques  are  very diversified  which  need to
be  handled  properly  for  further  investigation,  here  AI
based methods are very helpful for understanding the HIV
mechanism  far  better.  Diagnostic  and  therapeutic
techniques  available  for  HIV/AIDS  are  need  to  be
improved. This will bring  major  insights  for  beneficial of
mankind. Most importantly it would be very cost effective
and reachable to all the medical centers so nobody would
die in absence of treatment !

Corresponding author:
Dr.ANUBHA DUBEY EDUCATION DIRECTOR & TRAINER - KANISHKSOCIALMEDIA
(PHD,BIOINFORMATICS & MBA HR,)
PG,RESEARCH CONSALTANT IN CAREER CONSULING NATIONAL & INTERNATIONAL,GUIDANCE,UG,PG ETC. OUR MISSION: EDUCATION REVOLUTION,
 We provide Ph.D. advisory 1. Paper writing like a review, survey, communication, proposed method, etc. 2.Journal referencing, editing, publication. 3.Project proposal and writing. 4.Thesis writing and publication. As per the above service charges will apply.
 

(With input from news agency language)

If you like this story, share it with a friend! We are a non-profit organization. Help us financially to keep our journalism free from government and corporate pressure.
 

Post a Comment

0 Comments

Custom Real-Time Chart Widget

'; (function() { var dsq = document.createElement('script'); dsq.type = 'text/javascript'; dsq.async = true; dsq.src = '//' + disqus_shortname + '.disqus.com/embed.js'; (document.getElementsByTagName('head')[0] || document.getElementsByTagName('body')[0]).appendChild(dsq); })();

market stocks NSC