Deep Learning Can Predict Bevacizumab Therapeutic Effect and Microsatellite Instability Directly from Histology in Epithelial Ovarian Cancer.

Epithelial ovarian cancer (EOC) remains a significant cause of mortality among gynecologic cancers, with the majority of cases being diagnosed at an advanced stage. Before targeted therapies were available, EOC treatment relied largely on debulking surgery and platinum-based chemotherapy. Vascular endothelial growth factors have been identified as inducing tumor angiogenesis. According to several clinical trials, anti-vascular endothelial growth factor-targeted therapy with bevacizumab was effective in all phases of EOC treatment. However, there are currently no biomarkers accessible for regular therapeutic use despite the importance of patient selection. Microsatellite instability (MSI), caused by a deficiency of the DNA mismatch repair system, is a molecular abnormality observed in EOC associated with Lynch syndrome. Recent evidence suggests that angiogenesis and MSI are interconnected. Developing predictive biomarkers, which enable the selection of patients who might benefit from bevacizumab-targeted therapy or immunotherapy, is critical for realizing personalized precision medicine. In this study, we developed 2 improved deep learning methods that eliminate the need for laborious detailed image-wise annotations by pathologists and compared them with 3 state-of-the-art methods to not only predict the efficacy of bevacizumab in patients with EOC using mismatch repair protein immunostained tissue microarrays but also predict MSI status directly from histopathologic images. In prediction of therapeutic outcomes, the 2 proposed methods achieved excellent performance by obtaining the highest mean sensitivity and specificity score using MSH2 or MSH6 markers and outperformed 3 state-of-the-art deep learning methods. Moreover, both statistical analysis results, using Cox proportional hazards model analysis and Kaplan-Meier progression-free survival analysis, confirm that the 2 proposed methods successfully differentiate patients with positive therapeutic effects and lower cancer recurrence rates from patients experiencing disease progression after treatment (P < .01). In prediction of MSI status directly from histopathology images, our proposed method also achieved a decent performance in terms of mean sensitivity and specificity score even for imbalanced data sets for both internal validation using tissue microarrays from the local hospital and external validation using whole section slides from The Cancer Genome Atlas archive.

Using Arterial Pulse and Laser Doppler Analyses to Discriminate between the Cardiovascular Effects of Different Running Levels.

BACKGROUND AND AIMS: Running can induce advantageous cardiovascular effects such as improved arterial stiffness and blood-supply perfusion. However, the differences between the vascular and blood-flow perfusion conditions under different levels of endurance-running performance remains unclear. The present study aimed to assess the vascular and blood-flow perfusion conditions among 3 groups (44 male volunteers) according to the time taken to run 3 km: Level 1, Level 2, and Level 3. METHODS: The radial blood pressure waveform (BPW), finger photoplethygraphy (PPG), and skin-surface laser-Doppler flowmetry (LDF) signals of the subjects were measured. Frequency-domain analysis was applied to BPW and PPG signals; time- and frequency-domain analyses were applied to LDF signals. RESULTS: Pulse waveform and LDF indices differed significantly among the three groups. These could be used to evaluate the advantageous cardiovascular effects provided by long-term endurance-running training, such as vessel relaxation (pulse waveform indices), improvement in blood supply perfusion (LDF indices), and changes in cardiovascular regulation activities (pulse and LDF variability indices). Using the relative changes in pulse-effect indices, we achieved almost perfect discrimination between Level 3 and Level 2 (AUC = 0.878). Furthermore, the present pulse waveform analysis could also be used to discriminate between the Level-1 and Level-2 groups. CONCLUSIONS: The present findings contribute to the development of a noninvasive, easy-to-use, and objective evaluation technique for the cardiovascular benefits of prolonged endurance-running training.

Annotation-Free Deep Learning-Based Prediction of Thyroid Molecular Cancer Biomarker BRAF (V600E) from Cytological Slides.

Thyroid cancer is the most common endocrine cancer. Papillary thyroid cancer (PTC) is the most prevalent form of malignancy among all thyroid cancers arising from follicular cells. Fine needle aspiration cytology (FNAC) is a non-invasive method regarded as the most cost-effective and accurate diagnostic method of choice in diagnosing PTC. Identification of BRAF (V600E) mutation in thyroid neoplasia may be beneficial because it is specific for malignancy, implies a worse prognosis, and is the target for selective BRAF inhibitors. To the authors' best knowledge, this is the first automated precision oncology framework effectively predict BRAF (V600E) immunostaining result in thyroidectomy specimen directly from Papanicolaou-stained thyroid fine-needle aspiration cytology and ThinPrep cytological slides, which is helpful for novel targeted therapies and prognosis prediction. The proposed deep learning (DL) framework is evaluated on a dataset of 118 whole slide images. The results show that the proposed DL-based technique achieves an accuracy of 87%, a precision of 94%, a sensitivity of 91%, a specificity of 71% and a mean of sensitivity and specificity at 81% and outperformed three state-of-the-art deep learning approaches. This study demonstrates the feasibility of DL-based prediction of critical molecular features in cytological slides, which not only aid in accurate diagnosis but also provide useful information in guiding clinical decision-making in patients with thyroid cancer. With the accumulation of data and the continuous advancement of technology, the performance of DL systems is expected to be improved in the near future. Therefore, we expect that DL can provide a cost-effective and time-effective alternative tool for patients in the era of precision oncology.

Compliance with Oral Nutritional Supplementation among Gastric Cancer Patients at Nutritional Risk: A Cross-Sectional Study.

AIMS: To explore compliance with oral nutritional supplementation (ONS) and to identify the risk factors for noncompliance among gastric cancer patients based on the health belief model (HBM). METHODS: This prospective, observational study included gastric cancer patients at nutritional risk who were prescribed ONS from July to September 2020. Demographic factors, clinical factors, ONS-related factors, social factors and variables derived from the HBM were collected. The outcome of interest was compliance with ONS, which was measured by self-reported intake of ONS. Uni- and multivariate analyses of potential risk factors for noncompliance were performed. RESULTS: A total of 162 gastric cancer patients in the preoperative and adjuvant chemotherapy periods were analyzed. The compliance rate with ONS was 24.7%. Univariate analysis identified thirteen variables as risk factors for decreased compliance. Multivariate logistic analysis indicated that ONS compliance was independently associated with the treatment period, perceived barriers to ONS, the motivation to take ONS, and the timing of taking ONS. CONCLUSION: This study showed that overall ONS compliance among gastric cancer patients was notably low. Patients in the chemotherapy treatment period who took ONS at random times each day perceived more barriers to taking ONS and had a lower level of motivation were associated with lower compliance with ONS.

The effect of giving verbal feedback during neck stabilisation exercise as an addition to physical therapy in patients with chronic neck pain: A randomised controlled trial.

OBJECTIVES: To study the addition of feedback-guided neck strength home exercise to physical therapy as an enhanced rehabilitation programme in the treatment of patients with chronic neck pain. DESIGN: A prospective randomised controlled trial. SETTING: Rehabilitation department of an academic hospital. SUBJECTS: Patients with chronic neck pain. INTERVENTIONS: The patients in both groups received supervised physical therapy sessions 3 times a week for 12 weeks. Patients in Group A (N = 38) used the neck strengthening exerciser device for 20 min daily at home for 6 weeks and patients in Group B (N = 20) performed 20 min of daily regular neck exercise at home for 6 weeks. OUTCOME MEASURES: Neck disability index, pain visual analogue scale, active range of motion of the neck, Patient Global Assessment and patient evaluation of treatment effect. All subjects were assessed at baseline as well as at 6- and 12-week follow-ups. RESULTS: At the 6-week follow-up, Group A exhibited significantly greater improvements (P < 0.05) in pain Visual Analogue Scale (Group A: 2.97 ± 1.57; Group B: 4.20 ± 1.82), neck disability index (Group A: 13.95 ± 8.07; Group B: 20.07 ± 9.14) and active cervical extension (Group A: 65.26 ± 12.76; Group B: 51.45 ± 11.78). At 12-week follow-up, Group A also exhibited significantly greater active cervical extension (Group A: 67.74 ± 11.94; Group B: 53.85 ± 14.09; P < 0.05). CONCLUSION: Adding neck strengthening exerciser home training to physical therapy was demonstrated to be more effective than physical therapy alone for patients with chronic neck pain.

Rapid Detection of Fusarium oxysporum Using Insulated Isothermal PCR and a Rapid, Simple DNA Preparation Protocol.

We developed an insulated isothermal PCR (iiPCR) method for the efficient and rapid detection of Fusarium oxysporum (Fo), which is a fungus that infects various hosts and causes severe crop losses. The Fo iiPCR method was sensitive enough to detect up to 100 copies of standard DNA template and 10 fg of Fo genomic DNA. In addition, it could directly detect 1 pg of mycelium and 10 spores of Fo without DNA extraction. Our study compared the performance of Fo iiPCR to that of three published in planta molecular detection methods-conventional PCR, SYBR green-based real-time PCR, and hydrolysis probe-based real-time PCR-in field detection of Fo. All diseased field samples yielded positive detection results with high reproducibility when subjected to an Fo iiPCR test combined with a rapid DNA extraction protocol compared to Fo iiPCR with an automated magnetic bead-based DNA extraction protocol. Intraday and interday assays were performed to ensure the stability of this new rapid detection method. The results of detection of Fo in diseased banana pseudostem samples demonstrated that this new rapid detection method was suitable for field diagnosis of Fusarium wilt and had high F1 scores for detection (the harmonic mean of precision and recall of detection) for all asymptomatic and symptomatic Fo-infected banana samples. In addition, banana samples at four growth stages (seedling, vegetative, flowering and fruiting, and harvesting) with mild symptoms also showed positive detection results. These results indicate that this new rapid detection method is a potentially efficient procedure for on-site detection of Fo.

Comparison of the corticosteroid injection and hyaluronate in the treatment of chronic subacromial bursitis: A randomized controlled trial.

OBJECTIVE: To analyse the effectiveness of corticosteroid (CS) and hyaluronic acid (HA) subacromial - subdeltoid (SASD) injection compared with normal saline (NS) in patients with chronic subacromial bursitis (CSB). DESIGN: A prospective three-arm double-blinded randomised controlled trial. SETTING: Rehabilitation department of two teaching hospitals. SUBJECTS: Patients with CSB (N = 186) divided into CS (N = 68), HA (N = 60), and NS (N = 58) groups. INTERVENTIONS: Three SASD injections under ultrasound guidance: group A, 20 mg of triamcinolone; group B, 2.5 mL of HA; and group C, 2.5 mL of NS. OUTCOME MEASURES: The primary outcome measures were the pain visual analogue scale (VAS) score at eight weeks. The secondary outcomes were scores on the Shoulder Pain and Disability Index (SPADI) and Shoulder Disability Questionnaire. RESULTS: At eight weeks, the pain VAS scores during activity were 2.56 ± 2.29, 3.65 ± 2.50, and 4.71 ± 2.83 in the CS, HA, and NS groups, respectively (CS vs NS, P < 0.001; HA vs NS, P = 0.013; CS vs HA, P = 0.010). SPADI scores were 40.83 ± 21.75, 36.92 ± 22.78, and 33.35 ± 23.38 in the CS, HA, and NS groups, respectively (CS vs NS, P < 0.001; HA vs NS, P = 0.197; CS vs HA, P = 0.004). CONCLUSION: Ultrasound-guided corticosteroid injection into the subacromial - subdeltoid bursa was proven to be effective and superior to hyaluronic acid and normal saline injection for treating CSB. Hyaluronic acid injection was only marginally more effective than normal saline injection.Trial Registration: ClinicalTrials.gov: NCT02702206.

First report of anthracnose caused by Colletotrichum fructicola on tea in Taiwan.

Tea (Camellia sinensis (L.) O. Kuntze) is a very popular beverage and cash crop that is widely cultivated in tropical and subtropical areas. In November 2017, diseased tea plants that exhibiting brown blight disease were observed in Guanxi Township of Hsinchu County in Taiwan. In the plantation,15% of tea trees (about 4000 plants) had an average of 20% of the leaves with at least one lesion. The symptoms began as small, water-soaked lesions on young leaves and twigs and later became larger, dark brown, necrotic lesions of 1 to 3 cm in diameter on leaves and 2 to 5 cm in length on twigs. Symptomatic leaf tissue (1 cm2) from five samples per sample) was surface sterilized with 1% NaClO (from commercial bleach, Clorox) for 1 min, washed with sterilized water 3 times, plated onto potato dextrose agar (PDA), and incubated under 12h/12h cycles of light and darkness at 25°C until sporulation to determine the causal agent. A fungus was consistently isolated from symptomatic leaf samples (80% isolation rate). The fungus initially produced white-to-gray fluffy aerial hyphae, which subsequently exhibited dark pigmentation. Acervuli and setae were absent. The conidia were hyaline, aseptate, smooth-walled, and cylindrical with obtuse to slightly rounded ends, with sizes of 12.10 to 16.02 × 3.58 to 4.91 (average 13.77 × 4.05, n = 30) µm. The majority had two rounded guttules. The appressoria were brown to dark brown, ovoid and slightly obtuse at the tip in shape, had lengths ranging from 3.59 to 10.31 µm (with an average of 7.18 µm, n = 30), and had diameters of 3.14 to 6.43 µm (with an average of 5.10 µm, n = 30). Morphological characteristics matched the descriptions of Colletotrichum fructicola (Liu et al. 2015; Fuentes-Aragón et al. 2018). The internal transcribed spacer of nuclear ribosomal DNA (ITS), actin (ACT), chitin synthase (CHS-1), and Apn2-Mat1-2 intergenic spacer and partial mating-type Mat1-2 gene (ApMAT) sequences of the isolates were obtained to confirm this identification. The sequences showed close identity with those of C. fructicola ex-type cultures ICMP18581 and CBS 130416 (Weir et al. 2012) of 99.65% for the ITS (JX010165), 99.29% for the ACT (JX009501), and 100.00% for the CHS-1 (JX009866), as well as close identity with the other ex-type culture LF506 (Liu et al. 2015) of 99.59% for the ApMat (KJ954567), supporting the isolate's identification as C. fructicola. The sequences were deposited in GenBank, with the following accession Nos.: MN608177 (ITS), MN393175 (ACT), MT087546 (CHS-1), and MT087542 (ApMAT). Based on morphology and DNA sequence analysis, the associated fungus was identified as C. fructicola. Pathogenicity tests were performed next according to the procedures described in Chen et al. (2017). Healthy leaves on tea plants (Ca. sinensis 'Chin-shin Oolong') were wounded by pinpricking in the middle of each counterpart and inoculated with conidial suspension (1 × 107 conidia/ml, 10 µl). Both non-wounded and wounded healthy leaves were inoculated with the conidial suspension and sterile distilled water (a water control). The tea plants were covered with plastic bags to maintain high relative humidity for two days. One week after inoculation, anthracnose was observed on 40% of inoculated leaves, whereas all the control leaves remained healthy. The fungus was re-isolated from the diseased plants, and identified as C. fructicola by resequencing of the four genes. To the best of our knowledge, this is the first report of anthracnose caused by C. fructicola on tea in Taiwan although the pathogen has been present in China and Indonesia (Wang et al. 2016; Shi et al. 2017; Farr and Rossman, 2020).

The Use of Wearable Sensors for the Movement Assessment on Muscle Contraction Sequences in Post-Stroke Patients during Sit-to-Stand.

Electromyography (EMG) sensors have been used to study the sequence of muscle contractions during sit-to-stand (STS) in post-stroke patients. However, the majority of the studies used wired sensors with a limited number of placements. Using the latest improved wearable technology with 16 sensors, the current study was a thorough investigation to evaluate the contraction sequences of eight key muscles on the trunk and bilateral limbs during STS in post-stroke patients, as it became feasible. Multiple wearable sensors for the detection of muscle contraction sequences showed that the post-stroke patients performed STS with abnormal firing sequences, not only in the primary mover on the sagittal plane during raising, but also in the tibialis anterior, which may affect anticipatory postural adjustment in the gluteus medius, which may affect balance control. The abnormal tibialis anterior contraction until the early ascending phase and the delayed firing of the gluteus muscles highlight the importance of whole-kinetic-chain monitoring of contraction sequences using wearable sensors. The findings can be helpful for the design of therapeutic exercises.

The Design and Application of Simplified Insole-Based Prototypes with Plantar Pressure Measurement for Fast Screening of Flat-Foot.

This study aimed to find the correlation between conventional Arch Index (AI) measurements and our prototype of a simplified insole-based plantar pressure measurement system and to find out the effective plantar pressure sensor position. Twenty-one subjects participated in this study, which was divided into two groups: 10 subjects with flatfoot and 11 subjects with normal foot. Five force sensitive resistance sensors were used on this prototype using Arduino as the data acquisition device. Two types of trials, namely static and dynamic, were conducted to validate our system against the ink-type AI measurement as a golden standard. The results showed that in the static trial, there was a high linear correlation with the medial arch sensor configuration, while in the dynamic trial, there was a high linear correlation in the medial arch sensor configuration and sensor 5 configuration. This study showed that both static and dynamic tests using the self-developed device could effectively determine most of the flatfoot subjects and suggests that in the future, it can be applied in clinical applications because of its advantages when compared to the expensive-high tech graphic input board and conventional tools, like ink-type based measurements.

Multiple-Wearable-Sensor-Based Gait Classification and Analysis in Patients with Neurological Disorders.

The aim of this study was to conduct a comprehensive analysis of the placement of multiple wearable sensors for the purpose of analyzing and classifying the gaits of patients with neurological disorders. Seven inertial measurement unit (IMU) sensors were placed at seven locations: the lower back (L5) and both sides of the thigh, distal tibia (shank), and foot. The 20 subjects selected to participate in this study were separated into two groups: stroke patients (11) and patients with neurological disorders other than stroke (brain concussion, spinal injury, or brain hemorrhage) (9). The temporal parameters of gait were calculated using a wearable device, and various features and sensor configurations were examined to establish the ideal accuracy for classifying different groups. A comparison of the various methods and features for classifying the three groups revealed that a combination of time domain and gait temporal feature-based classification with the Multilayer Perceptron (MLP) algorithm outperformed the other methods of feature-based classification. The classification results of different sensor placements revealed that the sensor placed on the shank achieved higher accuracy than the other sensor placements (L5, foot, and thigh). The placement-based classification of the shank sensor achieved 89.13% testing accuracy with the Decision Tree (DT) classifier algorithm. The results of this study indicate that the wearable IMU device is capable of differentiating between the gait patterns of healthy patients, patients with stroke, and patients with other neurological disorders. Moreover, the most favorable results were reported for the classification that used the combination of time domain and gait temporal features as the model input and the shank location for sensor placement.

Immediate and long-term efficacy of laterally-wedged insoles on persons with bilateral medial knee osteoarthritis during walking.

BACKGROUND: The current study aimed to investigate the immediate and long-term effects of laterally-wedged (LW) insoles on the knee loadings, the knee abductor moment (KAM) in particular, and the compensatory changes at other lower limb joints in patients with bilateral medial knee osteoarthritis during level walking with and without LW insoles. METHODS: Older adults with bilateral medial knee OA (age 66 ± 5.3 years; height 156 ± 4.9 cm; mass 60 ± 5.1 kg; leg length 83.72 ± 3.64 cm) were studied using computerized gait analysis initially (Baseline) and 6 weeks after using LW insoles (Follow-up) during barefoot walking and walking with LW insoles (7° of lateral inclination, with medial arch support). The three-dimensional angles and internal moments at the lower limb joints, as well as the ground reaction forces, were obtained using a motion analysis system and two forceplates. Key features of all the variables were compared using paired t tests for immediate effects (barefoot vs. LW) and for long-term effects (Baseline vs. Follow-up). The symptomatic severity (WOMAC Index) was also evaluated (Baseline vs. Follow-up). RESULTS: The KAM with LW insoles at Baseline was significantly reduced when compared to the barefoot condition (p < 0.05), suggesting that the LW insoles were effective in reducing unfavorable loadings at the knee immediately upon wearing the insoles. After 6 weeks of wearing LW insoles (Follow-up), no significant changes were found in most of the biomechanical variables, including KAM (p > 0.05), when compared to Baseline with LW insoles. However, a specific gait adaptation with reduced knee loading was revealed when walking without LW insoles, i.e., for the barefoot condition (p < 0.05). CONCLUSIONS: After long-term use of LW insoles, the pain and physical function were improved with decreased peak KAM. A specific gait adaptation with reduced KAM was also found when walking without LW insoles. These results indicate a positive long-term effect in persons with bilateral medial knee OA, both as an orthosis to assist walking, and as a treatment intervention to facilitate gait adaptations in favor of reduced KAM.

Interactive boxing-cycling on frailty and activity limitations in frail and prefrail older adults: A randomized controlled trial.

BACKGROUND: Frailty is common among older adults, often associated with activity limitations during physical and walking tasks. The interactive boxing-cycling combination has the potential to be an innovative and efficient training method, and our hypothesis was that interactive boxing-cycling would be superior to stationary cycling in improving frailty and activity limitations in frail and prefrail older adults. OBJECTIVE: To examine the impact of interactive boxing-cycling on frailty and activity limitations in frail and prefrail older adults compared to stationary cycling. MATERIALS AND METHODS: A single-blinded randomized controlled trial. Forty-five participants who met at least one frailty phenotype criteria were randomly assigned to receive either interactive boxing-cycling (n = 23) or stationary-cycling (n = 22) for 36 sessions over 12 weeks. The interactive boxing-cycling was performed on a cycle boxer bike with an interactive boxing panel fixed in front of the bike. The primary outcomes were frailty status, including score and phenotypes. Secondary outcomes included activity limitations during physical and walking tasks. The pre- and post-intervention data of both groups were analyzed using a repeated measures two-way ANOVA. RESULTS: Both types of cycling significantly improved frailty scores (p<0.001). Interactive boxing-cycling was more effective than stationary cycling in reversing the frailty phenotype of muscle weakness (p = 0.03, odds ratio 9.19) and demonstrated greater improvements than stationary cycling in arm curl (p = 0.002, η2=0.20), functional reach (p = 0.001, η2=0.22), and grip strength (p = 0.02, η2=0.12) tests. Additionally, interactive boxing-cycling exhibited a greater effect on gait speed (p = 0.02, η2=0.13) and gait variability (p = 0.01, η2=0.14) during dual-task walking. CONCLUSION: In frail and prefrail older adults, interactive boxing-cycling effectively improves frailty but is not superior to stationary cycling. However, it is more effective at improving certain activity limitations. REGISTRATION NUMBER: TCTR20220328001.

Comparison of intra-articular hyaluronic acid injections with transcutaneous electric nerve stimulation for the management of knee osteoarthritis: a randomized controlled trial.

OBJECTIVE: To compare the effects of intra-articular hyaluronic acid (HA; ARTZ) and transcutaneous electric nerve stimulation (TENS) in the treatment of patients with knee osteoarthritis. DESIGN: A prospective, randomized controlled trial. SETTING: Rehabilitation clinic of a teaching hospital. PARTICIPANTS: Patients with knee osteoarthritis (N=50; aged 51-80y) were randomly assigned to the HA group (n=27) or the TENS group (n=23). INTERVENTIONS: The HA group received intra-articular HA injection into the affected knee once a week for 5 consecutive weeks, and the TENS group received a 20- minute session of TENS 3 times a week for 4 consecutive weeks. MAIN OUTCOME MEASURES: The primary outcome measures used were the visual analog scale (VAS) for pain and the Lequesne index. The secondary outcome measures were range of motion of the knee, walking time, pain threshold, patient global assessment, and disability in activities of daily living. All subjects were assessed at baseline, and at 2 weeks, 2 months, and 3 months after the treatments were completed. RESULTS: The TENS group exhibited a significantly greater improvement in VAS than the HA group at 2 weeks' follow-up (4.17 ± 1.98 vs 5.31 ± 1.78, respectively; P=.03). In addition, the TENS group also exhibited a significantly greater improvement in the Lequesne index than the HA group at 2 weeks' follow-up (7.78 ± 2.08 vs 9.85 ± 3.54, respectively; P=.01) and at 3 months' follow-up (7.07 ± 2.85 vs 9.24 ± 4.04, respectively; P=.03). CONCLUSIONS: TENS with silver spike point electrodes was observed to be more effective than intra-articular HA injection for patients with knee osteoarthritis in improving the VAS for pain at 2 weeks' follow-up as well as the Lequesne index at 2 weeks' and 3 months' follow-up.

Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target Therapy.

Breast cancer is the leading cause of cancer-related deaths among women worldwide, and early detection and treatment has been shown to significantly reduce fatality rates from severe illness. Moreover, determination of the human epidermal growth factor receptor-2 (HER2) gene amplification by Fluorescence in situ hybridization (FISH) and Dual in situ hybridization (DISH) is critical for the selection of appropriate breast cancer patients for HER2-targeted therapy. However, visual examination of microscopy is time-consuming, subjective and poorly reproducible due to high inter-observer variability among pathologists and cytopathologists. The lack of consistency in identifying carcinoma-like nuclei has led to divergences in the calculation of sensitivity and specificity. This manuscript introduces a highly efficient deep learning method with low computing cost. The experimental results demonstrate that the proposed framework achieves high precision and recall on three essential clinical applications, including breast cancer diagnosis and human epidermal receptor factor 2 (HER2) amplification detection on FISH and DISH slides for HER2 target therapy. Furthermore, the proposed method outperforms the majority of the benchmark methods in terms of IoU by a significant margin (p<0.001) on three essential clinical applications. Importantly, run time analysis shows that the proposed method obtains excellent segmentation results with notably reduced time for Artificial intelligence (AI) training (16.93%), AI inference (17.25%) and memory usage (18.52%), making the proposed framework feasible for practical clinical usage.

Weakly supervised bilayer convolutional network in segmentation of HER2 related cells to guide HER2 targeted therapies.

Overexpression of human epidermal growth factor receptor 2 (HER2/ERBB2) is identified as a prognostic marker in metastatic breast cancer and a predictor to determine the effects of ERBB2-targeted drugs. Accurate ERBB2 testing is essential in determining the optimal treatment for metastatic breast cancer patients. Brightfield dual in situ hybridization (DISH) was recently authorized by the United States Food and Drug Administration for the assessment of ERRB2 overexpression, which however is a challenging task due to a variety of reasons. Firstly, the presence of touching clustered and overlapping cells render it difficult for segmentation of individual HER2 related cells, which must contain both ERBB2 and CEN17 signals. Secondly, the fuzzy cell boundaries make the localization of each HER2 related cell challenging. Thirdly, variation in the appearance of HER2 related cells is large. Fourthly, as manual annotations are usually made on targets with high confidence, causing sparsely labeled data with some unlabeled HER2 related cells defined as background, this will seriously confuse fully supervised AI learning and cause poor model outcomes. To deal with all issues mentioned above, we propose a two-stage weakly supervised deep learning framework for accurate and robust assessment of ERBB2 overexpression. The effectiveness and robustness of the proposed deep learning framework is evaluated on two DISH datasets acquired at two different magnifications. The experimental results demonstrate that the proposed deep learning framework achieves an accuracy of 96.78 ± 1.25, precision of 97.77 ± 3.09, recall of 84.86 ± 5.83 and Dice Index of 90.77 ± 4.1 and an accuracy of 96.43 ± 2.67, precision of 97.82 ± 3.99, recall of 87.14 ± 10.17 and Dice Index of 91.87 ± 6.51 for segmentation of ERBB2 overexpression on the two experimental datasets, respectively. Furthermore, the proposed deep learning framework outperforms 15 state-of-the-art benchmarked methods by a significant margin (P<0.05) with respect to IoU on both datasets.

Detection of ERBB2 and CEN17 signals in fluorescent in situ hybridization and dual in situ hybridization for guiding breast cancer HER2 target therapy.

The overexpression of the human epidermal growth factor receptor 2 (HER2) is a predictive biomarker in therapeutic effects for metastatic breast cancer. Accurate HER2 testing is critical for determining the most suitable treatment for patients. Fluorescent in situ hybridization (FISH) and dual in situ hybridization (DISH) have been recognized as FDA-approved methods to determine HER2 overexpression. However, analysis of HER2 overexpression is challenging. Firstly, the boundaries of cells are often unclear and blurry, with large variations in cell shapes and signals, making it challenging to identify the precise areas of HER2-related cells. Secondly, the use of sparsely labeled data, where some unlabeled HER2-related cells are classified as background, can significantly confuse fully supervised AI learning and result in unsatisfactory model outcomes. In this study, we present a weakly supervised Cascade R-CNN (W-CRCNN) model to automatically detect HER2 overexpression in HER2 DISH and FISH images acquired from clinical breast cancer samples. The experimental results demonstrate that the proposed W-CRCNN achieves excellent results in identification of HER2 amplification in three datasets, including two DISH datasets and a FISH dataset. For the FISH dataset, the proposed W-CRCNN achieves an accuracy of 0.970±0.022, precision of 0.974±0.028, recall of 0.917±0.065, F1-score of 0.943±0.042 and Jaccard Index of 0.899±0.073. For DISH datasets, the proposed W-CRCNN achieves an accuracy of 0.971±0.024, precision of 0.969±0.015, recall of 0.925±0.020, F1-score of 0.947±0.036 and Jaccard Index of 0.884±0.103 for dataset 1, and an accuracy of 0.978±0.011, precision of 0.975±0.011, recall of 0.918±0.038, F1-score of 0.946±0.030 and Jaccard Index of 0.884±0.052 for dataset 2, respectively. In comparison with the benchmark methods, the proposed W-CRCNN significantly outperforms all the benchmark approaches in identification of HER2 overexpression in FISH and DISH datasets (p<0.05). With the high degree of accuracy, precision and recall , the results show that the proposed method in DISH analysis for assessment of HER2 overexpression in breast cancer patients has significant potential to assist precision medicine.

Interpretable attention-based deep learning ensemble for personalized ovarian cancer treatment without manual annotations.

Inhibition of pathological angiogenesis has become one of the first FDA approved targeted therapies widely tested in anti-cancer treatment, i.e. VEGF-targeting monoclonal antibody bevacizumab, in combination with chemotherapy for frontline and maintenance therapy for women with newly diagnosed ovarian cancer. Identification of the best predictive biomarkers of bevacizumab response is necessary in order to select patients most likely to benefit from this therapy. Hence, this study investigates the protein expression patterns on immunohistochemical whole slide images of three angiogenesis related proteins, including Vascular endothelial growth factor, Angiopoietin 2 and Pyruvate kinase isoform M2, and develops an interpretable and annotation-free attention based deep learning ensemble framework to predict the bevacizumab therapeutic effect on patients with epithelial ovarian cancer or peritoneal serous papillary carcinoma using tissue microarrays (TMAs). In evaluation with five-fold cross validation, the proposed ensemble model using the protein expressions of both Pyruvate kinase isoform M2 and Angiopoietin 2 achieves a notably high F-score (0.99±0.02), accuracy (0.99±0.03), precision (0.99±0.02), recall (0.99±0.02) and AUC (1.00±0). Kaplan-Meier progression free survival analysis confirms that the proposed ensemble is able to identify patients in the predictive therapeutic sensitive group with low cancer recurrence (p<0.001), and the Cox proportional hazards model analysis further confirms the above statement (p=0.012). In conclusion, the experimental results demonstrate that the proposed ensemble model using the protein expressions of both Pyruvate kinase isoform M2 and Angiopoietin 2 can assist treatment planning of bevacizumab targeted therapy for patients with ovarian cancer.

Increased hip adductor activation during sit-to-stand improves muscle activation timing and rising-up mechanics in individuals with hemiparesis.

Long sit-to-stand (STS) time has been identified as a feature of impaired functional mobility. The changes in biomechanics of STS performance with simultaneous hip adductor contraction have not been studied, which may limit indications for use of hip adductor activation during STS training. Ten individuals with hemiplegia (mean age 61.8 years, injury time 29.8 ± 15.2 months) performed the STS with and without squeezing a ball between two legs. The joint moments, ground reaction force (GRF), chair reaction force and movement durations and temporal index of electromyography were calculated from the control condition for comparison with those from the ball squeezing condition. Under the squeeze condition, reduced peak vertical GRF during the ascension phase with increased loading rate was observed in the nonparetic limb, and the peak knee extensor moment occurred earlier in the paretic. Earlier activation of tibialis anterior and gluteus maximus, and gluteus medius were found in squeeze STS. Squeezing a ball between limbs during STS increased the contraction timing of tibialis anterior, gluteus maximus, gluteus medius, and soleus as well as a more symmetric rising mechanics encourage the use of squeezing a ball between limbs during STS for individuals with hemiparesis.

Learning more from the inter-rater reliability of interstitial fibrosis assessment beyond just a statistic.

Interstitial fibrosis assessment by renal pathologists lacks good agreement, and we aimed to investigate its hidden properties and infer possible clinical impact. Fifty kidney biopsies were assessed by 9 renal pathologists and evaluated by intraclass correlation coefficients (ICCs) and kappa statistics. Probabilities of pathologists' assessments that would deviate far from true values were derived from quadratic regression and multilayer perceptron nonlinear regression. Likely causes of variation in interstitial fibrosis assessment were investigated. Possible misclassification rates were inferred on reported large cohorts. We found inter-rater reliabilities ranged from poor to good (ICCs 0.48 to 0.90), and pathologists' assessments had the worst agreements when the extent of interstitial fibrosis was moderate. 33.5% of pathologists' assessments were expected to deviate far from the true values. Variation in interstitial fibrosis assessment was found to be correlated with variation in interstitial inflammation assessment (r2 = 32.1%). Taking IgA nephropathy as an example, the Oxford T scores for interstitial fibrosis were expected to be misclassified in 21.9% of patients. This study demonstrated the complexity of the inter-rater reliability of interstitial fibrosis assessment, and our proposed approaches discovered previously unknown properties in pathologists' practice and inferred a possible clinical impact on patients.