Transcript John Leyendecker, M.D., Vice Chair of Clinical Operations, Professor

June 29, 2016 Presentation to the
Interagency Working Group on
Medical Imaging
John R. Leyendecker, M.D.
Vice Chairman of Clinical Operations
and
Professor of Radiology
UT Southwestern Medical Center
Dallas, TX
Who am I?
• I am a practicing Radiologist with 23 years experience.
– Vascular and interventional radiology (USAF)
– Abdominal imaging and intervention
• UTSA, Wake Forest, UT Southwestern
• My time is spent:
–
–
–
–
Clinical (30%)
Educational (10%)
Research (10%)
Administrative (50%)
• I receive no federal research funding.
But I will be a patient some day!
Outline
• Why focus on imaging?
• A tale of two patients: Present care versus
future care.
• How do we get there from here?
Detection
Diagnosis
treatment
response
Localization and
severity assessment
Treatment
Treatment planning
Imagine fighting a war in which you…
can’t find the enemy
can’t determine the enemy’s strength
have limited means to deliver weapons to the battlefield
have no idea whether your weapons are effective
We all share the same
challenges
•
•
•
•
•
•
Image processing and enhancement
Feature extraction
Noise reduction
Data compression
Data mining/Machine learning
Information systems integration and
data sharing
“What frustrates you that you wish you could
change?”
A tale of two patients
WHERE WE ARE NOW
WHERE WE CAN BE
A 75 year old man has blood in his urine detected at a routine
physical exam. His primary care doctor refers him to a urologist
• The urologist accesses the
patient’s electronic medical
record to order an imaging
exam.
•
The urologist accesses the patient’s
electronic medical record to order
an imaging exam.
• Cumulative radiation dose
• Allergies
• Implanted devices
• Systems integration and data sharing
Documentation regarding
the
• Evidence-based
decision support
patient’s allergy history and CKD
are buried deep in the medical
record and go unnoticed.
• The order is placed for a
contrast-enhanced CT
Warning: This patient has had a prior
severe reaction to iodinated contrast
and grade 3 CKD
click here for details
Based on the information you entered and
published evidence, the following imaging
tests are considered appropriate and do
not require pre-approval
Ultrasound of kidneys
MRI of kidneys
WHERE WE ARE NOW
The patient arrives for his CT scan.
Although the scan is completed, he
has a reaction to the intravenous
contrast material necessitating
resuscitation and an overnight
admission to a nearby hospital. The
prolonged hypotensive episode
further worsens his renal function.
WHERE WE CAN BE
Assessment of the kidneys is
performed on a high field open MRI
scanner using rapid free-breathing
sequences that generate a portfolio of
reproducible and standardized
quantitative measurements displayed
as parametric maps overlaying high
spatial resolution anatomic images.
• Technological innovation
• Standardization and quantification
T1= 800 msec
T2 = 24 msec
R2* = 22
ADC = 0.923
Ve = 10 mL/100mL
Vp = 9 mL/100mL
Fat content = 0%
WHERE WE ARE NOW
WHERE WE CAN BE
• Computer Assisted Functions identify
and volumetrically measure a mass in
the kidney.
• Segmentation software estimates the
risk of surgical resection based on
lesion size and location
A mass is identified in the •kidney
and
Data/image
analysis(nephrometry).
software
measured manually by the• radiologist
Databases correlating imaging
in a single axial dimension. phenotypes with genomics
and
• The quantitative
MRI data is digitally
outcomescompared to an extensive open-
access national database of renal
tumors, correlating the tumor’s
imaging “fingerprint” with biomarkers
associated with specific genetic
mutations, biologic behavior, and
molecular targets.
WHERE WE ARE NOW
• The busy radiologist quickly looks
through the 800 acquired images for
evidence of metastatic disease.
WHERE WE CAN BE
• Image data is integrated with
information available in the patient’s
portable medical database to determine
a risk/benefit profile for various
treatment strategies.
• Software surveys the rest of the image
data for additional findings and lesions
• A small lung nodule is missed.
• Image analysis software
demonstrating a similar molecular
Seamlessofintegration ofsignature.
imaging,
• She doesn’t mention the• indicators
metabolic syndrome, such clinical,
as fatty and risk stratification data
• It detects a 6 mm lung nodule. Software
liver disease, or the coronary artery
automatically calculates a risk profile for
calcifications, because that takes
the nodule.
additional time and she knows the
urologist won’t follow-up on those
• Software also determines risk profile
findings
for diabetes, heart disease, and
osteoporosis-related fractures.
Name: Bob
National medical record number: 33333
Age: 75
Lifetime effective medical radiation dose: 23 mSv
Implanted devices: none
Allergies: iodinated contrast material
Contrast administered: 8 ml brand X
Complications: none
Findings requiring follow-up
Finding
Method
Interval
Lungs: normal
6 mm pulmonary
Chest CT
6 months
Critical FindingsLiver: normal
nodule
Gallbladder: normal
Based on your patient’s risk factors, the risk of malignancy is 8%
Right renal normal
mass
Spleen:
Click here to enroll your patient in a lung nodule registry and clinic
 99% probability of Renal Cell Carcinoma
 97% probability
of type I papillary type
Pancreas:
normal
 Stage T1a
Adrenal
glands:
normal
 Nephrometry
score = 4p
Additional Findings
Kidneys: a 2.0 cm solid enhancing mass is seen in the right
• Hepatic steatosis (fatty liver) Click here for more information
kidney
• Colonic diverticulosis Click here for more information
Bowel:
normal
Coronal MR image
Musculoskeletal: normal
Risk profile
Lymph nodes:
normal
Mutated genes
Imaging-based
Diabetes
Mod
Click here for more information
genomic analysis
Other: NoMET
free air, no free fluid
Imaging features suggest a
significant likelihood that mutations
in the following genes are present
SETD2
BAP1
Ischemic heart
disease
Mod
NASH
Mod
Click here for more information
Impression: Enhancing mass in the right
kidney concerning
OsteoporosisLow
Click here for more information
related fracture
for renal cell carcinoma.
Oncologic follow-up
Right renal mass
5-30-2020
Location
Upper pole, right kidney
Volume
35 ml
Metastasis #1
N/A
Metastasis #2
N/A
1st follow-up
Click here for more information
2nd follow-up
Click here to view all images
Click here for a 3D printed model
Click here for interactive virtual resection
WHERE WE ARE NOW
• The urologist performs a partial
nephrectomy in the operating room
under general anesthesia, because
that’s what the urologist does for all
small renal masses.
WHERE WE CAN BE
•
Evidence-based outcomes data is
combined with patient-specific data to
determine the relative risk of disease
progression, complications, and cost vis• Pathology report: Renal cell carcinoma,
à-vis various treatment strategies.
papillary type, T1a.
• Outcomes data
• Cost effectiveness data
• The patient undergoes repeat imaging
• His cancer qualifies as low risk for
every 6 months .
progression. Surveillance has the highest
area under risk/benefit/cost curve of all
• The patient’s renal function never fully
possible treatment strategies.
recovers from his contrast reaction and
deteriorates further after surgery,
• Based on this data, the patient chooses
necessitating dialysis.
annual imaging surveillance.
WHERE WE ARE NOW
WHERE WE CAN BE
The patient is referred back to
his primary doctor for follow-up
of his lung nodule and for
lifestyle and medical
interventions for type II diabetes,
mild renal insufficiency, fatty
liver disease, and coronary artery
disease.
The patient develops
complications related to
poorly controlled
diabetes/metabolic syndrome
and renal insufficiency and
spends his final days in and
out of hospitals
He dies a short time later from
a heart attack.
He spends many happy years
enjoying the company of his
grandchildren.
How do we get there from here?
STANDARDIZATION
• Image acquisition
• Image analysis
• Image reporting
Support and Encourage…
• Continued technological innovation
• Faster, safer, more effective
• Vendor neutral quantitative imaging
• Data management and integrated information systems
• Large databases with clinical trials data
• Machine learning and computer aided diagnostic tools
• Outcomes data (also emphasizing when NOT to treat)
• Contrast agents with better safety profiles and targeted
ligands for diagnosis, therapy, and surveillance.
• Minimally invasive procedures
• Insertion of novel imaging into therapeutic clinical trials
• Potential to accelerate clinical translation
Core contrast
Glycosylated liposome
Build a research infrastructure that
fosters innovation
• We can’t just invest in tools, we also need to invest in people
and processes.
• Create the next generation of radiologist (translational)
scientists
– Debt, lack of time, lack of training are all barriers
• Develop and encourage successful industry/investigator
collaborative models
• Revamp the grants review process
– Should every grant be treated like an R01?
– Reinvigorate the R21/R33 grant?