Search for: All records

Heterogeneities among tumor cells significantly contribute towards cancer progression and therapeutic inefficiency. Herein, we discuss recent microfluidic platforms for sorting and profiling of tumor cells for prognostics and personalized therapies. 

Microwave reflection photoconductive decay carrier lifetimes of Ge0.94Sn0.06 materials on oriented GaAs substrates at 300 K. 

Germanium alloyed with α-tin (GeSn) transitions to a direct bandgap semiconductor of significance for optoelectronics. It is essential to localize the carriers within the active region for improving the quantum efficiency in a GeSn based laser. In this work, epitaxial GeSn heterostructure material systems were analyzed to determine the band offsets for carrier confinement: (i) a 0.53% compressively strained Ge 0.97 Sn 0.03 /AlAs; (ii) a 0.81% compressively strained Ge 0.94 Sn 0.06 /Ge; and (iii) a lattice matched Ge 0.94 Sn 0.06 /In 0.12 Al 0.88 As. The phonon modes in GeSn alloys were studied using Raman spectroscopy as a function of Sn composition, that showed Sn induced red shifts in wavenumbers of the Ge–Ge longitudinal optical phonon mode peaks. The material parameter b representing strain contribution to Raman shifts of a Ge 0.94 Sn 0.06 alloy was determined as b = 314.81 ± 14 cm −1 . Low temperature photoluminescence measurements were performed at 79 K to determine direct and indirect energy bandgaps of E g,Γ = 0.72 eV and E g,L = 0.66 eV for 0.81% compressively strained Ge 0.94 Sn 0.06 , and E g,Γ = 0.73 eV and E g,L = 0.68 eV for lattice matched Ge 0.94 Sn 0.06 epilayers. Chemical effects of Sn atomic species were analyzed using X-ray photoelectron spectroscopy (XPS), revealing a shift in Ge 3d core level (CL) spectra towards the lower binding energy affecting the bonding environment. Large valence band offset of Δ E V = 0.91 ± 0.1 eV and conduction band offset of Δ E C,Γ–X = 0.64 ± 0.1 eV were determined from the Ge 0.94 Sn 0.06 /In 0.12 Al 0.88 As heterostructure using CL spectra by XPS measurements. The evaluated band offset was found to be of type-I configuration, needed for carrier confinement in a laser. In addition, these band offset values were compared with the first-principles-based calculated Ge/InAlAs band alignment, and it was found to have arsenic up-diffusion limited to 1 monolayer of epitaxial GeSn overlayer, ruling out the possibility of defects induced modification of band alignment. Furthermore, this lattice matched GeSn/InAlAs heterostructure band offset values were significantly higher than GeSn grown on group IV buffer/substrates. Therefore, a lattice matched GeSn/InAlAs material system has large band offsets offering superior carrier confinement to realize a highly efficient GeSn based photonic device. 

Cancer diagnostics is an important field of cancer recovery and survival with many expensive procedures needed to administer the correct treatment. Machine Learning (ML) approaches can help with the diagnostic prediction from circulating tumor cells in liquid biopsy or from a primary tumor in solid biopsy. After predicting the metastatic potential from a deep learning model, doctors in a clinical setting can administer a safe and correct treatment for a specific patient. This paper investigates the use of deep convolutional neural networks for predicting a specific cancer cell line as a tool for label free identification. Specifically, deep learning strategies for weight initialization and performance metrics are described, with transfer learning and the accuracy metric utilized in this work. The equipment used for prediction involves brightfield microscopy without the use of chemical labels, advanced instruments, or time-consuming biological techniques, giving an advantage over current diagnostic methods. In the procedure, three different binary datasets of well-known cancer cell lines were collected, each having a difference in metastatic potential. Two different classification models were adopted (EfficientNetV2 and ResNet-50) with the analysis given for each stage in the ML architecture. The training results for each model and dataset are provided and systematically compared. We found that the test set accuracy showed favorable performance for both ML models with EfficientNetV2 accuracy reaching up to 99%. These test results allowed EfficientNetV2 to outperform ResNet-50 at an average percent increase of 3.5% for each dataset. The high accuracy obtained from the predictions demonstrates that the system can be retrained on a large-scale clinical dataset.