For the busy clinical-imaging professional in an AI world: Gaining intuition about deep learning without math.

Medical diagnostics comprise recognizing patterns in images, tissue slides, and symptoms. Deep learning algorithms (DLs) are well suited to such tasks, but they are black boxes in various ways. To explain DL Computer-Aided Diagnostic (CAD) results and their accuracy to patients, to manage or drive the direction of future medical DLs, to make better decisions with CAD, etc., clinical professionals may benefit from hands-on, under-the-hood lessons about medical DL. For those who already have some high-level knowledge about DL, the next step is to gain a more-fundamental understanding of DLs, which may help illuminate inside the boxes. The objectives of this Continuing Medical Education (CME) article include:Better understanding can come from relatable medical analogies and personally experiencing quick simulations to observe deep learning in action, akin to the way clinicians are trained to perform other tasks. We developed readily-implementable demonstrations and simulation exercises. We framed the exercises using analogies to breast cancer, malignancy and cancer stage as example diagnostic applications. The simulations revealed a nuanced relationship between DL output accuracy and the quantity and nature of the data. The simulation results provided lessons-learned and implications for the clinical world. Although we focused on DLs for diagnosis, they are similar to DL CME. .

Spin-Filtering Ferroelectric Tunnel Junctions as Multiferroic Synapses for Neuromorphic Computing.

As nanoelectronic synapses, memristive ferroelectric tunnel junctions (FTJs) have triggered great interest due to the potential applications in neuromorphic computing for emulating biological brains. Here, we demonstrate multiferroic FTJ synapses based on the ferroelectric modulation of spin-filtering BaTiO3/CoFe2O4 composite barriers. Continuous conductance change with an ON/OFF current ratio of ∼54 400% and long-term memory with the spike-timing-dependent plasticity (STDP) of synaptic weight for Hebbian learning are achieved by controlling the polarization switching of BaTiO3. Supervised learning simulations adopting the STDP results as database for weight training are performed on a crossbar neural network and exhibit a high accuracy rate above 97% for recognition. The polarization switching also alters the band alignment of CoFe2O4 barrier relative to the electrodes, giving rise to the change of tunneling magnetoresistance ratio by about 10 times and even the reversal of its sign depending upon the resistance states. These results, especially the electrically switchable spin polarization, provide a new approach toward multiferroic neuromorphic devices with energy-efficient electrical manipulations through potential barrier design. In addition, the availability of spinel ferrite barriers epitaxially grown with ferroelectric oxides also expends the playground of FTJ devices for a broad scope of applications.

Virtual Learning Simulations in High School: Effects on Cognitive and Non-cognitive Outcomes and Implications on the Development of STEM Academic and Career Choice.

The present study compared the value of using a virtual learning simulation compared to traditional lessons on the topic of evolution, and investigated if the virtual learning simulation could serve as a catalyst for STEM academic and career development, based on social cognitive career theory. The investigation was conducted using a crossover repeated measures design based on a sample of 128 high school biology/biotech students. The results showed that the virtual learning simulation increased knowledge of evolution significantly, compared to the traditional lesson. No significant differences between the simulation and lesson were found in their ability to increase the non-cognitive measures. Both interventions increased self-efficacy significantly, and none of them had a significant effect on motivation. In addition, the results showed that the simulation increased interest in biology related tasks, but not outcome expectations. The findings suggest that virtual learning simulations are at least as efficient in enhancing learning and self-efficacy as traditional lessons, and high schools can thus use them as supplementary educational methods. In addition, the findings indicate that virtual learning simulations may be a useful tool in enhancing student's interest in and goals toward STEM related careers.

Simulation based virtual learning environment in medical genetics counseling: an example of bridging the gap between theory and practice in medical education.

BACKGROUND: Simulation based learning environments are designed to improve the quality of medical education by allowing students to interact with patients, diagnostic laboratory procedures, and patient data in a virtual environment. However, few studies have evaluated whether simulation based learning environments increase students' knowledge, intrinsic motivation, and self-efficacy, and help them generalize from laboratory analyses to clinical practice and health decision-making. METHODS: An entire class of 300 University of Copenhagen first-year undergraduate students, most with a major in medicine, received a 2-h training session in a simulation based learning environment. The main outcomes were pre- to post- changes in knowledge, intrinsic motivation, and self-efficacy, together with post-intervention evaluation of the effect of the simulation on student understanding of everyday clinical practice were demonstrated. RESULTS: Knowledge (Cohen's d = 0.73), intrinsic motivation (d = 0.24), and self-efficacy (d = 0.46) significantly increased from the pre- to post-test. Low knowledge students showed the greatest increases in knowledge (d = 3.35) and self-efficacy (d = 0.61), but a non-significant increase in intrinsic motivation (d = 0.22). The medium and high knowledge students showed significant increases in knowledge (d = 1.45 and 0.36, respectively), motivation (d = 0.22 and 0.31), and self-efficacy (d = 0.36 and 0.52, respectively). Additionally, 90 % of students reported a greater understanding of medical genetics, 82 % thought that medical genetics was more interesting, 93 % indicated that they were more interested and motivated, and had gained confidence by having experienced working on a case story that resembled the real working situation of a doctor, and 78 % indicated that they would feel more confident counseling a patient after the simulation. CONCLUSIONS: The simulation based learning environment increased students' learning, intrinsic motivation, and self-efficacy (although the strength of these effects differed depending on their pre-test knowledge), and increased the perceived relevance of medical educational activities. The results suggest that simulations can help future generations of doctors transfer new understanding of disease mechanisms gained in virtual laboratory settings into everyday clinical practice.