Explainable machine learning approach for cancer prediction through binarilization of RNA sequencing data.

In recent years, researchers have proven the effectiveness and speediness of machine learning-based cancer diagnosis models. However, it is difficult to explain the results generated by machine learning models, especially ones that utilized complex high-dimensional data like RNA sequencing data. In this study, we propose the binarilization technique as a novel way to treat RNA sequencing data and used it to construct explainable cancer prediction models. We tested our proposed data processing technique on five different models, namely neural network, random forest, xgboost, support vector machine, and decision tree, using four cancer datasets collected from the National Cancer Institute Genomic Data Commons. Since our datasets are imbalanced, we evaluated the performance of all models using metrics designed for imbalance performance like geometric mean, Matthews correlation coefficient, F-Measure, and area under the receiver operating characteristic curve. Our approach showed comparative performance while relying on less features. Additionally, we demonstrated that data binarilization offers higher explainability by revealing how each feature affects the prediction. These results demonstrate the potential of data binarilization technique in improving the performance and explainability of RNA sequencing based cancer prediction models.

Effects of Nd: YAG LASER irradiation and O2 plasma on the adhesive performance of poly-ether-ether-ketone (PEEK).

PURPOSE: To evaluate the effects of neodymium-doped yttrium aluminum garnet (Nd: YAG) LASER irradiation and oxygen (O2) plasma on the adhesive performance of polyether ether ketone (PEEK) and resin adhesive. METHODS: Nd: YAG LASERs of varying powers and O2 plasma for different durations were used to modify PEEK. A total of 168 PEEK specimens were randomly divided into seven groups (n = 24/group): (A) Control group: untreated PEEK, (B) L0.75 group: PEEK modified with 0.75 W Nd: YAG LASER, (C) L1 group: PEEK modified with 1.0 W Nd: YAG LASER, (D) L1.25 group: PEEK modified with 1.25 W Nd: YAG LASER, (E) P15 group: PEEK modified with 15 min of O2 plasma, (F) P25 group: PEEK modified with 25 min of O2 plasma, and (G) P35 group: PEEK modified with 35 min of O2 plasma. The surface characteristics of the materials were comprehensively analyzed using a scanning electron microscope (SEM), profilometer, energy-dispersive spectrometer (EDS), and contact angle tester. The adhesive specimens were bonded with Variolink N resin adhesive in all groups and each group was further divided into two subgroups (n = 12/group): (a) water storage for 56 h at 37 °C and (b) thermal cycling 5000 times. Shear bond strength (SBS) was tested using a universal testing machine, and the fracture modes were observed using an automated chemiluminescence analysis system to assess the effects of Nd: YAG LASER and O2 plasma on the bond strength of PEEK to resin adhesive. RESULTS: Both Nd: YAG LASER and O2 plasma treatments altered the surface characteristics of PEEK and significantly increased the SBS between PEEK and Variolink N resin adhesive. The L0.75 group (Nd: YAG LASER) and the P35 group (O2 plasma) achieved the highest SBS, respectively. Furthermore, the SBS of the L0.75 group was higher than that of the P35 group. Following thermal cycling, SBS values decreased compared to the water storage subgroups. The fracture modes of the specimens in each group were predominantly interfacial and mixed, with no cohesive fractures observed. CONCLUSIONS: Nd: YAG LASER irradiation and O2 plasma treatments can improve the SBS between PEEK and resin adhesive, with the 0.75 W Nd: YAG LASER being the preferred treatment method.

Synergistic Effect of Nano Strontium Titanate Coating and Ultraviolet C Photofunctionalization on Osteogenic Performance and Soft Tissue Sealing of poly(ether-ether-ketone).

This study aimed to evaluate the bioactivity of poly(ether ether ketone) (PEEK) after surface modification by persistent photoconductive strontium titanate (SrTiO3) magnetron sputtering and ultraviolet (UV) C irradiation. According to the different modifications, the PEEK specimens were randomly divided into five groups (n = 38/group): PEEK, Sr100-PEEK, Sr200-PEEK, UV/PEEK, and UV/Sr200-PEEK. Then, the specimens of Sr100-PEEK and Sr200-PEEK groups were, respectively, coated with 100 and 200 nm thickness photocatalyst SrTiO3 on the PEEK surface by magnetron sputtering. Subsequently, UV-C light photofunctionalized the specimens of PEEK and Sr200-PEEK groups to form UV/PEEK and UV/Sr200-PEEK groups. The specimens were characterized by a step meter, scanning electron microscopy (SEM), atomic force microscopy (AFM), energy dispersive X-ray spectroscopy (EDX), and a water contact angle meter. The release test of the Sr ion was performed by inductively coupled plasma mass spectrometry (ICP-MS). In vitro study, osteogenic activity (MC3T3-E1 osteoblast-like cells) and epithelial and connective tissue attachment (gingival epithelial cells GE1 and fibroblasts NIH3T3) were analyzed in five groups. Surface morphology of the specimens was changed after coating, and the Sr content on the Sr-PEEK surface was increased with increasing coating thickness. In addition, the contact angle was increased significantly after magnetron sputtering. After UV-C photofunctionalization, the content of surface elements changed and the contact angle was decreased. The release of Sr ion was sustained, and the final cumulative release amount did not exceed the safety limit. In vitro experiments showed that SrTiO3 improved the cell activity of MC3T3-E1 and UV-C irradiation further enhanced the osteogenic performance of PEEK. Besides, UV-C irradiation also significantly promoted the cell viability, development, and expression of adhesion proteins of GE1 and NIH3T3 on PEEK. The present investigation demonstrated that nano SrTiO3 coating with UV-C photofunctionalization synergistically enhanced the osteogenic properties and soft tissue sealing function of PEEK in vitro.