The winner of the Google Science Fair this year was a 17 year old girl who built a cloud-based service that helps doctors detect malignant breast cancer tumors 99% of the time. 
Medical usage demands neural networks achieve accuracy with their diagnosis and reduce malignant false negatives. Building on data collected by the University of Wisconsin in the early 1990s, this project first evaluates three modern commercial neural network implementations. Information regarding potential indicators of breast cancer is quantified in the dataset; specifically, clump thickness, single epithelial cell size, bare nuclei, mitoses, and five other attributes. Each network accepts this input to optimize its hidden nodes and is tested with ten trials. With each trial, a randomly selected 10% of the dataset is used to assess the predictive power of the constructed neural network. These commercially created neural networks serve as a control group.
Development of a custom neural network weights malignant false negatives and allows for the identification of inconclusive samples (capacities not available in commercial products.) Additionally, more samples are needed to improve the predictive capability of the network; therefore, the network has been published in the cloud, allowing for global submissions and benefit. The cloud service is hosted in the Google App Engine.
The successfully implemented custom network is tested with 6,800 trials. To assure maximum training, each sample is run through ten trials evaluated by different networks trained against all other samples. The custom neural network achieved predictive success of 97.4% with 99.1% sensitivity to malignancy – substantially better than the evaluated commercial products. Out of the commercial products, two experienced consistent success while the third experienced erratic success. The sensitivity to malignancy for the custom network was 5% higher than the best commercial network’s sensitivity. This experiment demonstrates modern neural networks can handle outliers and work with unmodified datasets to identify patterns. In addition, when all data is used for training, the custom network achieves 100% success with only 4 inconclusive samples, proving the network is more effective with more samples. Additionally, 7.6 million trials were run using different training sample sizes to demonstrate the sensitivity and predictive success improves as the network receives more training samples.
The Global Neural Network Cloud Service for Breast Cancer may be ready to diagnose actual patients – more global participation is required to confirm the findings and increase the predictive success on blind samples.
(Editor’s note: I weep with joy knowing that the world’s problems are being tackled and solved by increasingly younger minds.)
Dave Chappelle on the miracle of children.
YMFY is for the children. Please like this baby photo on Facebook so we can win this silly baby contest. Thanks!
Women and children first
For decades the science of child-rearing was guided by patriarchal ideas, but now the cradle rocks to an older rhythm. Eric Michael Johnson, in conversation with eminent evolutionary biologists Sarah Hrdy and Robert Trivers, explores how Mother Nature and the social network that nurtured our past have been remembered at last.
Hrdy believes that for hundreds of thousands of years, mothers and children were given the physical and emotional support that allowed our species to thrive. Hunter-gatherers have always relied on a network of attachments so that, should one caregiver fail, many others could ensure emotionally confident and secure individuals.
“Rates of child mortality were high, but there was no child abuse or emotional neglect,” Hrdy told me. “A child that experienced the kind of emotional neglect it takes to produce the psychopathology of insecure attachment, the kind showed in Bowlby’s and Harlow’s research, simply would not have survived.”
“Never hug and kiss them, never let them sit on your lap. If you must, kiss them once on the forehead when they say good night. Shake hands with them in the morning.” Only now do we realise the full impact of 1920s parenting advice.
An environment that contained a network of support for mothers and children was formative in our species’ development. We have forgotten these memories today and, as a result, deceived ourselves about what children, and our society as a whole, ultimately need to feel secure.
I have never watched children die in front of me before. Watching their last breath as their chest slowly and with long pauses slightly expand and then deflate again. Until, it suddenly stops. The children who arrived at the Banadir hospital in Mogadishu were in bad shape, but they were the lucky ones. Some of them who made it to the hospital early enough managed to pull through, even with limited medical supplies and overworked, unpaid, and tired nurses. However, for most, it was a place they came to die. Almost all the children I photographed on the second floor in the children’s wing ended up dying. With some I did not even have a chance to know their names or ages. I would return to the room a couple of hours later and the bed the child was lying in before was either empty, or full again with a new child and mother. —Dominic Nahr, Magnum for TIME
Must Love Kids: What we talk about when we talk about a woman's success as a woman
By: Jessa Crispin
Excerpted from a review of Susan Hertog’s biography of two remarkable writers, Dangerous Ambition: Rebecca West and Dorothy Thompson
Holding onto our mother’s grudges isn’t just affecting our current or future children. It also has to make us snipey and judgey about how other women raise their children. It can make us take two powerful women and diminish them down to “failed mothers.” Perhaps the most feminist thing we can do is simply forgive our mothers whatever slights. Unless she was putting her cigarettes out on your arm when you were a kid, of course. If we can see the Good Enough Mother that Winnicott championed in our own flawed mothers, maybe we can see the same in the women around us. We can appreciatethem for something other than the products of their womb. Or, at least not let their specific missteps overshadow the rest of their lives.
Doing a little daytime drinking with a friend recently, the conversation circled around to our mothers. We went through the usual list of complaints, dredging up old hurts and comparing stories. My friend told me he had recently read that Spinoza, through his entire life, carried with him the bed in which his mother died. He carted it from residence to residence, sleeping every night literally in his own mother’s deathbed. Sure, it was a different world, before IKEA and constantly renewing mattress technology, but come on. He surely could have swapped or sold the bed, but he held onto it. And if Spinoza — fucking Spinoza, man! my friend said, shaking his head — couldn’t let go of his mother issues, then we must be doing all right, him and me.
There’s another side to that, of course: That if Spinoza — fucking Spinoza, man! — couldn’t get over his mother issues, then surely we are all doomed to the same fate, no matter how much psychotherapy or how many philosophical structures we go through. I didn’t say it out loud then, because really, there’s not enough alcohol in the world to make that all right.
