BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20211207T055411Z
LOCATION:223
DTSTART;TZID=America/Chicago:20211114T103000
DTEND;TZID=America/Chicago:20211114T104500
UID:submissions.supercomputing.org_SC21_sess434_ws_cafcw123@linklings.com
SUMMARY:An Integrated Simulation-HPC-Learning Approach to Create Cancer Pa
 tient Templates for Digital Twins
DESCRIPTION:Workshop\n\nAn Integrated Simulation-HPC-Learning Approach to 
 Create Cancer Patient Templates for Digital Twins\n\nLima da Rocha\n\nCanc
 er patient digital twins (CPDTs) are personalized simulation models that c
 an forecast individuals’ prognosis under a variety of treatment options. T
 o successfully launch CPDTs, we must combine mechanistic modeling, artific
 ial intelligence (AI), and high performance computing (HPC) into a platfor
 m that can seamlessly combine the patient’s data with accumulated knowledg
 e, while continuously learning from successes and failures. \n\nBuilding f
 rom recent COVID-19 modeling, we developed a multiscale agent-based model 
 of melanoma micrometastases and immune response in lung tissue. The cellul
 ar components of the tumor microenvironment are formed by healthy epitheli
 al cells from the lung, isolated melanoma cells, and immune cells: macroph
 ages, dendritic cells, CD8+, and CD4+ T cells. The model also includes imm
 une cell trafficking to and from the lymphatic system to drive an expandin
 g immune response. Melanoma cells proliferate uncontrolled, causing mechan
 ical stress in the region of the cell cluster. This mechanical factor lead
 s to the death of healthy cells in the region, stimulating the activation 
 of antigen-presenting cells (APCs). Apoptotic melanoma cell death can also
  activate APCs. Macrophages ingest dead cell debris and produce proinflamm
 atory cytokines, recruiting more APCs to the region. Dendritic cells are a
 ctivated when in contact with dying cells (lung and melanoma) and 'receive
 ' their antigen signature. Activated dendritic cells migrate to the lymph 
 node and recruit CD8+ and CD4+ T cells that can induce cancer cell death. 
 \n\nWe analyzed the parameter space by using high-throughput model explora
 tion on HPC to generate over 100k virtual patient trajectories, and demons
 trated that the model can recapitulate a broad variety of virtual patient 
 trajectories, including dormancy, uncontrolled growth, and partial and com
 plete tumor response. Moreover, we used AI techniques to cluster the virtu
 al trajectories into CPDT templates – the first step in fitting a personal
 ized model to an individual patient. To fit a personalized model to an ind
 ividual patient, we used several different bi-clustering techniques on CPD
 T patient templates and virtual trajectories. Our initial results suggest 
 that some CPDT patient profiles may not be distinguishable in the initial 
 stages, and that complete tumor elimination may represent a “lucky” stocha
 stic event in the broader population whose cancer is otherwise partially c
 ontrolled by the immune system, rather than a distinct subpopulation.\n\nT
 ag: Applications, Computational Science, Education and Training and Outrea
 ch, HPC Community Collaboration, HPC Training and Education, Machine Learn
 ing and Artificial Intelligence, Performance, Workforce\n\nRegistration Ca
 tegory: Workshop Reg Pass
END:VEVENT
END:VCALENDAR