Deep learning mammography model detects breast cancer up to five years in advance

MIT CSAIL professor Regina Barzilay and Harvard/MGH professor Constance Lehman  have developed a deep learning model that can predict breast cancer, from a mammogram, up to five years in the future. The model learned subtle breast tissue patterns that lead to malignant tumors from mammograms and known outcomes of 90,000 MGH patients.

The goal is to individualize screening and prevention programs.

Barzilay said that “rather than taking a one-size-fits-all approach, we can personalize screening around a woman’s risk of developing cancer.  For example, a doctor might recommend that one group of women get a mammogram every other year, while another higher-risk group might get supplemental MRI screening.”

The algortithm accurately placed 31 percent of all cancer patients in its highest-risk category, compared to 18 percent for traditional models.

Lehman hopes to change screening strategies from age-based to risk based. “This is because before we did not have accurate risk assessment tools that worked for individual women.”

Current risk assement,  based on age, family history of breast and ovarian cancer, hormonal and reproductive factors, and breast density, are weakly correlated with breast cancer. This makes many organizations believe that risk-based screening is not possible.

Rather than manually identifying the patterns in a mammogram that drive future cancer, the algorithm deduced patterns directly from the data, detecting abnormalities too subtle for the human eye to see.

Lehman said that “since the 1960s radiologists have noticed that women have unique and widely variable patterns of breast tissue visible on the mammogram. These patterns can represent the influence of genetics, hormones, pregnancy, lactation, diet, weight loss, and weight gain. We can now leverage this detailed information to be more precise in our risk assessment at the individual level.”

The MIT/MGH model  is equally accurate for white and black women, as opposed to prior models. Black women have been shown to be 42 percent more likely to die from breast cancer due to a wide range of factors that may include differences in detection and access to health care.

Barzilay believes the system could, in the future,  determine, based on mammograms, if patients are at a greater risk for cardiovascular disease or other cancers.


Join ApplySci at the 12th Wearable Tech + Digital Health + Neurotech Boston conference on November 14, 2019 at Harvard Medical School and the 13th Wearable Tech + Neurotech + Digital Health Silicon Valley conference on February 11-12, 2020 at Stanford University

Starving cancer stem cells as a new approach to glioblastoma

Luis Parada and Sloan Kettering colleagues are focusing on cancer stem cells as a new approach to glioblastoma.

Like normal stem cells, cancer stem cells have the ability to rebuild a tumor, even after most of it has been removed, leading to cancer relapse and metastasis.

According to Parada: “The pharmaceutical industry has traditionally used established cancer cell lines to screen for new drugs, but these cell lines don’t always reflect how cancer behaves in the body. The therapies that are currently in use were designed to target cells that are rapidly dividing. But what we’ve concluded in our studies is that glioblastoma stem cells divide relatively slowly within tumors, leaving them unaffected by these treatments.”

Even if most of the tumor is destroyed, the stem cells allow it to regrow.

The team discovered a drug, which they called Gboxin, that effectively treated glioblastoma in mice, and killed human glioblastoma cells.  They then discovered that Gboxin killed cancer stem cells by starving them of energy – . by preventing cells from making ATP through oxidative phosphorylation in mitochondria.  When Gboxin accumulates within cancer stem cells, it essentially strangles the mitochondria and shuts energy production down.

The next step is to determine that Gboxin will be able to cross the blood-brain barrier, and potential side effects of the drug.


Join ApplySci at the 12th Wearable Tech + Digital Health + Neurotech Boston conference on November 14, 2019 at Harvard Medical School and the 13th Wearable Tech + Neurotech + Digital Health Silicon Valley conference on February 11-12, 2020 at Stanford University

“Monorail” could halt spread of brain tumors

Duke’s Ravi Bellamkonda has developed a “Tumor Monorail” which tricks aggressive brain tumors such as glioblastoma into migrating into an external container rather than throughout the brain.  It has been designated “Breakthrough Device” by the U.S. Food and Drug Administration (FDA).

The device mimics the physical properties of the brain’s white matter to entice aggressive tumors to migrate toward the exterior of the brain, where the migrating cells can be collected and removed. It does not to destroy the tumor, but does halt its lethal spread. There are no chemicals or enzymes involved, and there are a wide variety of materials that the device could be made from.

The work is based on rat studies from 2014.  The team hopes to receive FDA approval for human trials by the end of 2019.

Click to view Georgia Tech (whose researchers collaborated with colleagues at Emory and Duke) video


Join ApplySci at the 10th Wearable Tech + Digital Health + Neurotech Silicon Valley conference on February 21-22 at Stanford University — Featuring:  Zhenan BaoChristof KochVinod KhoslaWalter Greenleaf – Nathan IntratorJohn MattisonDavid EaglemanUnity Stoakes Shahin Farshchi Emmanuel Mignot Michael Snyder Joe Wang – Josh Duyan – Aviad Hai Anne Andrews Tan Le – Anima Anandkumar – Pierrick Arnal – Shea Balish – Kareem Ayyad – Mehran Talebinejad – Liam Kaufman – Scott Barclay – Tracy Laabs – George Kouvas

Wireless system could track tumors, dispense medicine

Dina Katabi and MIT CSAIL colleagues have developed ReMix, which uses lo power wireless signals to pinponit the location of implants in the body.  The tiny implants could be used as tracking devices on shifting tumors to monitor  movements, and in the future to deliver drugs to specific regions.

The technology showed centimeter-level accuracy in animal tests.

Markers in the body reflect the signal transmitted by the wireless device outside the body, therefore a battery or external power source are not required.


Join ApplySci at the 9th Wearable Tech + Digital Health + Neurotech Boston conference on September 24, 2018 at the MIT Media Lab.  Speakers include:  Rudy Tanzi – Mary Lou Jepsen – George ChurchRoz PicardNathan IntratorKeith JohnsonJuan EnriquezJohn MattisonRoozbeh GhaffariPoppy Crum – Phillip Alvelda Marom Bikson – Ed Simcox – Sean Lane

AI – optimized glioblastoma chemotherapy

Pratik Shah, Gregory Yauney,  and MIT Media Lab researchers have developed an AI  model that could make glioblastoma chemotherapy regimens less toxic but still effective. It analyzes current regimens and iteratively adjusts doses to optimize treatment with the lowest possible potency and frequency toreduce tumor sizes.

In simulated trials of 50 patients, the machine-learning model designed treatment cycles that reduced the potency to a quarter or half of the doses It often skipped administration, which were then scheduled twice a year instead of monthly.

Reinforced learning was used to teach the model to favor certain behavior that lead to a desired outcome.  A combination of  temozolomide and procarbazine, lomustine, and vincristine, administered over weeks or months, were studied.

As the model explored the regimen, at each planned dosing interval it decided on actions. It either initiated or withheld a dose. If it administered, it then decided if the entire dose, or a portion, was necessary. It pinged another clinical model with each action to see if the the mean tumor diameter shrunk.

When full doses were given, the model was penalized, so it instead chose fewer, smaller doses. According to Shah, harmful actions were reduced to get to the desired outcome.

The J Crain Venter Institute’s Nicholas Schork said that the model offers a major improvement over the conventional “eye-balling” method of administering doses, observing how patients respond, and adjusting accordingly.


Join ApplySci at the 9th Wearable Tech + Digital Health + Neurotech Boston conference on September 24, 2018 at the MIT Media Lab.  Speakers include:  Rudy Tanzi – Mary Lou Jepsen – George ChurchRoz PicardNathan IntratorKeith JohnsonJuan EnriquezJohn MattisonRoozbeh GhaffariPoppy Crum – Phillip Alvelda Marom Bikson – Ed Simcox – Sean Lane

Hydrogen peroxide sensor to determine effective chemotherapy

MIT’s Hadley Sikes has developed a sensor that determines whether cancer cells respond to a particular type of chemotherapy by detecting hydrogen peroxide inside human cells.

The technology could help identify new cancer drugs that boost levels of hydrogen peroxide, which induces programmed cell death. The sensors could also be adapted to screen individual patients’ tumors to predict whether such drugs would be effective against them.


Join ApplySci at the 9th Wearable Tech + Digital Health + Neurotech Boston conference on September 24, 2018 at the MIT Media Lab.  Speakers include:  Rudy Tanzi – Mary Lou Jepsen – George ChurchRoz PicardNathan IntratorKeith JohnsonJuan EnriquezJohn MattisonRoozbeh GhaffariPoppy Crum – Phillip Alvelda Marom Bikson – Ed Simcox – Sean Lane

Urine test for cancer biomarkers

Minoru Sakairi and Hitachi scientists have developed a urine test for early cancer detection.

5,000 types of metabolites can be analyzed for cancer biomarkers in urine.  The team began a study three years ago, resulting in the identification of 30 metabolites that can be used to discriminate between healthy people and cancer patients.  Further validation studies will begin in September at Nagoya University.

According to Sakairi:  “For the comprehensive analysis of urine metabolites, we used a liquid chromatograph/mass spectrometer (LC/MS). Taking measurements with an LC/MS, and focusing on differences in the water-and fat-solubility of metabolites so as to optimize measurement conditions, we were able to detect over 1,300 metabolites in the urine samples. Using 30 biomarkers from among these, a look at their measured values for 15 cases each of breast cancer patients, colorectal cancer patients, and healthy subjects showed that we had made a breakthrough in being able to discriminate the difference between cancer and not cancer.”


Join ApplySci at the 9th Wearable Tech + Digital Health + Neurotech Boston conference – September 25, 2018 a the MIT media Lab

Remote photodynamic therapy targets inner-organ tumors

NUS researchers Zhang Yong and John Ho have developed a tumor-targeting method that remotely conveys light  for  photodynamic treatment.

The tiny, wireless, implanted device delivers doses of light over a long period  in a programmable and repeatable manner.

PDT is usually used on surface diseases because of  low infiltration of light through organic tissue. This remote approach to light conveyance allows PDT to be used on the inner organs with fine control.  The team believes that it could successfully treat brain and liver malignancies in the future, and allow therapies that could be tailored during the course of treatment.


Join ApplySci at Wearable Tech + Digital Health + Neurotech Silicon Valley on February 26-27, 2018 at Stanford University. Speakers include:  Vinod Khosla – Justin Sanchez – Brian Otis – Bryan Johnson – Zhenan Bao – Nathan Intrator – Carla Pugh – Jamshid Ghajar – Mark Kendall – Robert Greenberg – Darin Okuda – Jason Heikenfeld – Bob Knight – Phillip Alvelda – Paul Nuyujukian –  Peter Fischer – Tony Chahine – Shahin Farshchi – Ambar Bhattacharyya – Adam D’Augelli – Juan-Pablo Mas – Shreyas Shah– Walter Greenleaf – Jacobo Penide – David Sarno – Peter Fischer

Registration rates increase on February 2nd

 

AI detects bowel cancer in less than 1 second in small study

Yuichi Mori and Showa University colleagues haved used AI to identify bowel cancer by analyzing colonoscopy derived polyps in less than a second.

The  system compares a magnified view of a colorectal polyp with 30,000 endocytoscopic images. The researchers claimed  86% accuracy, based on a study of 300 polyps.

While further testing the technology, Mori said that the team will focus on creating a system that can automatically detect polyps.

Click to view Endoscopy Thieme video


Join ApplySci at Wearable Tech + Digital Health + Neurotech Silicon Valley on February 26-27, 2018 at Stanford University. Speakers include:  Vinod Khosla – Justin Sanchez – Brian Otis – Bryan Johnson – Zhenan Bao – Nathan Intrator – Carla Pugh – Jamshid Ghajar – Mark Kendall – Robert Greenberg – Darin Okuda – Jason Heikenfeld – Bob Knight – Phillip Alvelda