Special Techniques in Cytology
• The devices scan the entire cellular content of the
slide without attention lapses or missed fields of
• The devices apply a standard set of analyses in a uniform
fashion to all cells or FOVs on the slide. Theoretically,
computerized cell or FOV measurements should lack the
variability, or nonobjectivity, that is inherent in human
analyses and which is subject to a wide variety of biases.
Although not widely appreciated, human ability to
"measure" an object, such as a nuclear size, is affected by
the cell's local environment, and biases such as precon-
ceptions about the case diagnosis (e.g. if an observer al-
ready "thinks" that a case represents a dysplasia, percep-
tion of size is likely to fall within the bounds of known
"dysplasia criteria" regardless of the actual measured
• In the automated approach, in which slides are triaged
into cases requiring review and those not requiring review,
based on overall probability of abnormality on the slide,
categorization as a low probability "no review" slide im-
proves specificity of abnormal detection. Put another way,
if humans were allowed to screen this population of slides
having very low probability of abnormality, a "finding"
would more likely be a false, rather than a true, positive
• In the location-guided screening approach, in which
a limited number of "highest probability" areas of the
slide are viewed by the human screener, the increased
"signal to noise" ratio (meaning more abnormal cells
are presented per normal cell) improves screener vigi-
lance and hence sensitivity to the abnormal cells present.
This concept has been clearly shown in the literature
about other types of screening procedures, such as visual
screening for prohibited objects at airport checkpoints.
If the prohibited objects are more prevalent in the total
population of objects viewed, the sensitivity of detec-
tion increases.43 A decreased false-negative rate can be
expected with the use of this focused screening approach.
In addition, in a similar fashion to the previous con-
cept, when screeners do not view the "low probability"
areas of the slide, the likelihood of false-positive results
should be diminished.
Key features b y w h ic h com puterized autom ated screening
im proves accuracy are as follow s.
• The device scans the entire slide without attention lapses
or missed FOVs.
• The device applies a standard set of analyses in a uniform
fashion to all cells and FOVs.
• In slide-ranking devices, slides triaged to no manual
review increase specificity by decreasing false-positive
• In slide-ranking devices, increased screener vigilance in
high-probability cases increases sensitivity by decreasing
• When devices are utilized in combination with liquid-
based preparation, improved cell visualization is syner-
gistic with the above four points, leading to yet further
improvements in accuracy.
T he c o m b in a tio n o f liquid-based cytology p reparation tech-
niques w ith com puterized autom ated screening has the p otential
to have a synergistic effect o n accuracy. Better visu a liza tio n o f
cells in h e re n t in the liquid-based technique sho uld a llo w fo r
m ore accurate and efficient analyses b y com p uter devices in a
s im ila r fa shion to h u m a n observers. T his hypothesis w ill be p ut
to the test w h en data fro m th e clinical trials and independent
studies using these systems are presented below.
P ro d u c tivity im p rovem ents w ith autom ated cytology m ethods
and devices are as fo llow .
• Liquid-based cytology specimens can be screened by
humans and by automated devices at a faster speed than
conventional cytology specimens. Studies have shown that
productivity gains of about 20% for manual screening can
be expected once personnel are trained and have gained
• Devices that triage slides into categories requiring hu-
man review and those that do not, benefit from fewer
slides entering the human review process. At present, this
productivity gain can be as high as 25% based on USFDA
regulatory operating parameters, but it is as much as 50%
in use outside the United States.
• Location-guided screening devices requiring only limited
reviews of machine-selected fields substantially reduce
the average time spent screening a slide. At present, in the
currently approved Cytyc device, this productivity gain is
reported to be as high as 100%.45
Suffice to say th a t a u to m a tio n provides theoretic and, in
m an y cases, achieved im p rovem ents in accuracy, b o th sensitiv-
ity and specificity, as w e ll as p ro d u c tivity gains. Such achieve-
m ents are im p o rta n t in an era o f d eclining resources, in b o th
reim b ursem ent and technolog ist labor, and the increased threat
o f costly litig a tio n fo r inaccurate results.
Currently Available Automation Platforms
Cytyc ThinPrep Pap Test
T h e T h in P re p Pap Test (H o lo g ic , M a rlb o ro u g h , M A ) was the
firs t U S FD A -app roved m e th o d , receiving clearance in 1996.
C o lle c tio n o f a T h in P re p specim en requires th e use o f e ith e r
a b ro o m -typ e device o r th e c o m b in a tio n o f an endocervical
b rush and spatula. C ells are w ashed o ff th e s a m p lin g device(s)
in to a tra n s p o rt v ia l (Preservcyt, H o lo g ic ) c o n ta in in g a p ro -
p rie ta ry preservative m ix tu re th a t is m ethanol-b ased . T h e v ia l
is received in th e lab o rato ry, w h ere th e cell susp ension u n d e r-
goes h o m o g e n iz a tio n , and th e re fo re ra n d o m iz a tio n , b y a vo r-
te xin g cell d isp ersion m eth od . C ells are tra nsferred b y suc tion
to a filte r (Fig. 3 4 .2 ) . C e llu la rity o f th e specim en is c o n tro lle d
b y m o n ito rin g o f a pressure g rad ie nt across th e filte r. W h e n
th is pressure g rad ie nt reaches a d efined level, th e suc tion stops
and th e cells are transferred to a glass slid e b y b lo ttin g and the
a p p lic a tio n o f gentle pressure across th e filte r in th e op p osite
d ire c tio n . T h e fin a l re s u lt is a circle o f deposited cells m easur-
ing 20 m m in d ia m e te r (Fig. 3 4 .3 ) . T he re are tw o m od els o f
th e T h in P re p processing device, b o th o f w h ic h operate in a
fu n d a m e n ta lly s im ila r fa s h io n as described above; th e T 2 0 0 0
m od e l (Fig. 3 4 .4 ) is a single specim en lo a d device, and the
T 3 0 0 0 m od e l (Fig. 3 4 .5 ) is an a uto m a te d m u ltilo a d device
designed fo r h ig h -v o lu m e lab oratories.