Before performing radiation therapy, radiation oncologists first carefully review medical images of the patient to identify the gross tumour volume, the observable portion of the disease. They then design patient-specific clinical target volumes that include surrounding tissues, since these regions can hide cancerous cells and provide pathways for metastasis.
Known as contouring, this process establishes how much radiation a patient will receive and how it will be delivered. In the case of head and neck cancer, this is a particularly sensitive task due to the presence of vulnerable tissues in the vicinity.
Though it may sound straightforward, contouring clinical target volumes is quite subjective. A recent study from Utrecht University found wide variability in how trained physicians contoured the same patient’s computed tomography (CT) scan, leading some doctors to suggest high-risk clinical target volumes eight times larger than their colleagues.
This inter-physician variability is a problem for patients, who may be over- or under-dosed based on the doctor they work with. It is also a problem for determining best practices, so standards of care can emerge.
Recently, Carlos Cardenas, a graduate research assistant and PhD candidate at The University of Texas MD Anderson Cancer Center in Houston, Texas, and a team of researchers at MD Anderson, working under the supervision of Laurence Court with support from the National Institutes of Health, developed a new method for automating the contouring of high-risk clinical target volumes using artificial intelligence and deep neural networks.
Cardenas’ work focuses on translating a physician’s decision-making process into a computer program. "We have a lot of clinical data and radiation therapy treatment plan data at MD Anderson," he said. "If we think about the problem in a smart way, we can replicate the patterns that our physicians are using to treat specific types of tumours."
In their study, they analysed data from 52 oropharyngeal cancer patients who had been treated at MD Anderson between January 2006 to August 2010, and had previously had their gross tumour volumes and clinical tumour volumes contoured for their radiation therapy treatment.
Cardenas spent a lot of time observing the radiation oncology team at MD Anderson, which has one of the few teams of head and neck subspecialist oncologists in the world, trying to determine how they define the targets.
"For high-risk target volumes, a lot of times radiation oncologists use the existing gross tumour disease and apply a non-uniform distance margin based on the shape of the tumour and its adjacent tissues," Cardenas said. "We started by investigating this first, using simple distance vectors."
Cardenas began the project in 2015 and had quickly accumulated an unwieldy amount of data to analyse. He turned to deep learning as a way of mining that data and uncovering the unwritten rules guiding the experts’ decisions.
The deep learning algorithm he developed uses auto-encoders — a form of neural networks that can learn how to represent datasets — to identify and recreate physician contouring patterns.
The model uses the gross tumour volume and distance map information from surrounding anatomic structures as its inputs. It then classifies the data to identify voxels — three-dimensional pixels — that are part of the high-risk clinical target volumes. In oropharyngeal cancer cases, the head and neck are usually treated with different volumes for high, low and intermediate risk. The paper described automating the target for the high-risk areas. Additional forthcoming papers will describe the low and intermediate predictions.
In addition to potentially reducing inter-physician variability and allowing comparisons of outcomes in clinical trials, a tertiary advantage of the method is the speed and efficiency it offers. It takes a radiation oncologist two to four hours to determine clinical target volumes. At MD Anderson, this result is then peer reviewed by additional physicians to minimize the risk of missing the disease.
Using the Maverick supercomputer at the Texas Advanced Computing Center (TACC), they were able to produce clinical target volumes in under a minute. Training the system took the longest amount of time, but for that step too, TACC resources helped speed up the research significantly.
"If we were to do it on our local GPU [graphics processing unit], it would have taken two months," Cardenas said. "But we were able to parallelize the process and do the optimization on each patient by sending those paths to TACC and that’s where we found a lot of advantages by using the TACC system."

Texas Advanced Computing Centerwww.tacc.utexas.edu/-/an-ai-oncologist-to-help-cancer-patients-worldwide

In pregnancies complicated by foetal congenital heart disease (CHD), global placental perfusion was significantly decreased and regional variation of placental perfusion significantly increased as pregnancies progressed, findings that point to non-invasive imaging providing an early warning of placental dysfunction. A Children’s National Health System research team is thought to be the first to report non-invasive, whole placenta perfusion imaging in utero in a study.
According to the National Heart, Lung and Blood Institute, congenital heart defects are the most common type of birth defect, affecting 8 in 1,000 newborns. Early in pregnancy, the foetal heart follows a parallel developmental course as the placenta, which supplies the growing foetus with oxygen and nutrients while ferrying out waste products. The study authors write that placental arteries are dynamic during pregnancy, remodelling themselves to accommodate increased blood flow as foetuses undergo explosive growth spurts in later stages of pregnancy. If this crucial remodelling does not occur, the placenta may not supply sufficient oxygen and nutrients to the foetus, leading to foetal growth restriction or preeclampsia.
The research team led by Catherine Limperopoulos, Ph.D., enrolled 48 pregnant women who underwent at least one foetal magnetic resonance imaging (MRI) session during their second or third trimester of pregnancy. Thirty-one of the women were healthy volunteers whose mean gestational age was 30 weeks (range: 21 to 39 gestational weeks). Seventeen women were pregnant with foetuses diagnosed with CHD whose mean gestational age was 32 weeks (range: 22 to 38 gestational weeks).
The researchers used velocity-selective arterial spin labelling (VSASL), a powerful MRI technique that directly measures the rate of delivery of arterial blood to organs like the brain. ASL tracks water molecules within the blood as blood flows through arteries, eliminating the need to use a contrast agent. The team was able to distinguish the placenta perfusion contributions by the foetus and the mother.
“In pregnancies complicated by foetal CHD, global placental perfusion significantly decreased and regional variation of placental perfusion significantly increased with advancing gestational age,” says Zungho “Wesley” Zun, Ph.D., the study’s lead author.
“Just like the human brain, heart and kidneys─organs that can commandeer heightened blood flow when needed─the placenta may employ an auto-regulatory mechanism to optimize perfusion,” adds Limperopoulos, director of Children’s Developing Brain Research Laboratory and the study’s senior author. “The early increased global placental profusion in pregnancies complicated by CHD may represent an attempt to correct for insufficient foetal blood flow.”
The research team writes that the findings demonstrate that placental dysfunction due to CHD can be apparent as early as the second trimester of pregnancy using this imaging technology.
“The predictive value of VSASL imaging, which we continue to study, holds the promise of detecting dysfunction before placental abnormalities become irreversible,” Limperopoulos says.
Children’s National Health Systemchildrensnational.org/news-and-events/childrens-newsroom/2018/advanced-mri-can-detect-placental-perfusion-abnormalities-in-pregnancies-complicated-by-fetal-chd

New research has found treating an infected hip replacement in a single stage procedure may be as effective or better than the widely used two-stage procedure. To date no well-designed study has compared these procedures head-to-head to decide if one is better or if they achieve the same results. Hip replacement is a very common operation that is effective at providing pain relief and improving mobility, however, infection can sometimes occur following joint replacement. The findings have wide implications for orthopaedic surgery, the NHS, and health systems worldwide.
The research team, led by the University of Bristol, conducted a study that reviewed patient data from 44 studies to compare the effectiveness of the two types of surgery currently used to treat infections – one-stage and two-stage revisions.
In the two-stage procedure, the existing artificial joint is removed in one operation and the patient is treated for several months with antibiotics.  A new joint is then inserted in a second operation.  In the one-stage procedure, the artificial joint is removed along with all infected tissue and a new one inserted in the same operation.
The study found that the one-stage revision strategy is as good, if not better, as the two-stage strategy. The one-stage strategy may also be better suited for patients with certain types of infection or problems that were previously thought not to be appropriate for this type of surgery.
Dr Setor Kunutsor, Research Fellow from the Musculoskeletal Research Unit at the Bristol Medical School: (THS) and lead researcher, said: “For several decades, the two-stage procedure has been presumed to be more effective than the one-stage. However, it has disadvantages for patients such as having two major surgical procedures, significant pain and limited function between stages, long hospital stays, as well as high healthcare costs. The one-stage strategy has potential advantages for patients which include having only one major surgery, shorter time in hospital, reduced functional impairment, and is less expensive.
“When the research team analysed the collected data, the findings confirmed what we had suspected all along – the one-stage strategy may be as effective as, or better than the two-stage strategy.
Speaking about the study, Co-investigator and Senior Author Mr. Andrew Beswick, also a Research Fellow of the Musculoskeletal Research Unit at the Bristol Medical School: (THS), said: “Our research and the subsequent adoption of the one-stage strategy by surgeons and hospitals, could improve lives, prevent unnecessary deaths, and save money.”
Bristol Universitywww.bristol.ac.uk/news/2018/april/hip-replacement.html

Cardiovascular disease pervades Appalachia, yet many Appalachians live far from any heart and vascular specialist. Follow-up doctor’s visits in the weeks after cardiovascular surgery can involve hours-long drives down narrow, winding roads.
A recent study led by Albeir Mousa, a professor in the West Virginia University School of Medicine, suggests telemedicine may improve these patients’ satisfaction with their postoperative care as well as their quality of life.
With telemedicine, a healthcare provider can use a computer, tablet or other electronic device to remotely evaluate their patients’ symptoms, diagnose illnesses or injuries, and prescribe treatments. They can also field their patients’ questions.
The 30 participants in Mousa’s study were recovering from vascular surgery. In each case, the surgeon made an incision in the patient’s groin to access the arteries that needed rebuilding or rerouting. Whether the incisions healed without complications was the study’s focus.
Sixteen patients received tablets with Enform—a telemedicine app developed by TeleMed 2020 Inc.—that facilitated communication with nurses managing their care. As part of an in-home monitoring kit, patients also received thermometers, blood pressure cuffs, scales and devices to measure blood oxygen saturation levels.
Each day, patients who had been discharged from the hospital weighed themselves, took their temperature, measured their pulse and blood pressure, and determined their blood oxygen levels using the Enform app. They completed a wellness and symptom tracking quiz that included questions like “How is your pain today?” Each week they answered satisfaction and emotional wellness questions as well. These data, along with photos of the surgical incision sites that patients captured with the app—were made available to the patients’ care team.
Care managers, in turn, logged into the telemedicine platform daily to review the information patients had submitted from their homes. Cares managers received notifications of abnormalities, such as blood pressure spikes and fevers. Based on the information they gathered, the care managers intervened, answered patients’ questions about symptoms or wound care, called in prescriptions, scheduled appointments with physicians, and modified care plans based on consultations with the medical director.
Meanwhile, the other 14 participants had standard-of-care treatment. They received no monitoring equipment, tablet or telemedicine app.
After 30 days, the researchers made a number of comparisons between the two groups. For example, were wound infections more common in one group than the other? Did one group require more hospital readmissions? How did members of each group rate their own well-being? Were they happy with the postoperative care they received?
Hospital-readmission and wound-infection rates did not differ significantly between groups. The researchers attribute this fact to the study’s small sample size. But patients in the telemedicine group scored better on measures of their physical function, mental health and role limitations due to physical health problems. In addition, the vast majority of patients who used the app found it intuitive to use. Using a five-point scale to measure ease of use, 91 percent of patients gave it a score of 4 or 5. A similar percentage of patients said the app enriched the quality of care they received.
Likewise, the telemedicine patients’ scores on quality-of-life assessments surged more dramatically between the study’s beginning and end.
Patients assigned to the telemedicine group lived an average of 60 miles from their vascular care center. Almost a third of them lived more than 77 miles away and had to drive for two to three hours to get there.
“Telemedicine would save a lot of headache in Appalachia—in areas where people don’t even have the money to get in the car to get to the hospital,” said Mousa, who teaches surgery at the WVU Health Sciences Charleston Campus.
He envisions that, one day, patients will be able to download a cell phone app that provides these telemedicine services. That way, they won’t even need a tablet. “Each household has at least one cell phone, and most likely, it’s a smartphone.”
“You’re getting the same service,” he said, “but with a very minor hassle for the patient and the physician.”

West Virginia University
medicine.hsc.wvu.edu/News/Story?headline=wvu-researchers-find-telemedicine-may-increase-patients-satisfaction-with-their-medical-care

Nowadays, tumours inside the complex central nervous system remain one of the most challenging cancers to diagnose.
Different from conventional brain-imaging techniques, nearinfrared (NIR) fluorescence imaging demonstrates particular merits including being non-hazardous, offering fast feedback, and having higher sensitivity.
A research team led by Prof. ZHENG Hairong from the Shenzhen Institutes of Advanced Technology (SIAT) of the Chinese Academy of Sciences, in collaboration with Prof. LIU Bin from the University of Singapore, reported the first NIR-II fluorescent molecule with aggregation-induced-emission (AIE) characteristics for dual fluorescence and photoacoustic imaging.
Fluorescence imaging in the second NIR window (NIR-II), compared with the first NIR window (NIR-I), exhibits salient advantages of deeper penetration and higher spatiotemporal resolution, owing to further reduced photon scattering, absorption, and tissue autofluorescence in biological tissues.
Scientists designed a new donor-acceptor (D-A)-tailored NIR-II emissive AIE molecule, and formulated dots showed a high NIR-II fluorescence quantum yield up to 6.2%, owing to the intrinsic aggregation-induced emission nature of the designed molecule.
The AIE dots have been successfully used for dual NIR-II fluorescence and NIR-I photoacoustic imaging for precise non-invasive brain-tumour diagnosis. Based on the same dots, the experiments revealed that NIR-II fluorescence imaging showed a high resolution.
Meanwhile, NIR-I PA imaging intrinsically exhibited higher penetration depth than that of NIR-II fluorescence imaging, which allowed clear delineation of tumour depth in the brain.
The synergetic bimodal imaging with targeting c-RGD-decorated bright AIE nanoparticles showed precise brain-tumour diagnosis with good specificity and high sensitivity, which yielded a high S/B of 4.4 and accurately assessed the depth of tumour location inside brain tissue.
The study demonstrates the promise of NIR-II AIE molecules and their dots in dual NIR-II fluorescence and NIR-I photoacoustic imaging for precise brain cancer diagnostics.

EurekAlert
www.eurekalert.org/pub_releases/2018-06/caos-nmh060818.php