A whole-slide foundation model for digital pathology from real-world data.

Digital pathology poses unique computational challenges, as a standard gigapixel slide may comprise tens of thousands of image tiles1-3. Prior models have often resorted to subsampling a small portion of tiles for each slide, thus missing the important slide-level context4. Here we present Prov-GigaPath, a whole-slide pathology foundation model pretrained on 1.3 billion 256 × 256 pathology image tiles in 171,189 whole slides from Providence, a large US health network comprising 28 cancer centres. The slides originated from more than 30,000 patients covering 31 major tissue types. To pretrain Prov-GigaPath, we propose GigaPath, a novel vision transformer architecture for pretraining gigapixel pathology slides. To scale GigaPath for slide-level learning with tens of thousands of image tiles, GigaPath adapts the newly developed LongNet5 method to digital pathology. To evaluate Prov-GigaPath, we construct a digital pathology benchmark comprising 9 cancer subtyping tasks and 17 pathomics tasks, using both Providence and TCGA data6. With large-scale pretraining and ultra-large-context modelling, Prov-GigaPath attains state-of-the-art performance on 25 out of 26 tasks, with significant improvement over the second-best method on 18 tasks. We further demonstrate the potential of Prov-GigaPath on vision-language pretraining for pathology7,8 by incorporating the pathology reports. In sum, Prov-GigaPath is an open-weight foundation model that achieves state-of-the-art performance on various digital pathology tasks, demonstrating the importance of real-world data and whole-slide modelling.

Protocol for a single-blind randomized clinical trial to test the efficacy of bilateral transcranial magnetic stimulation on upper extremity motor function in patients recovering from stroke.

BACKGROUND: No consensus currently exists regarding the optimal protocol for repetitive transcranial magnetic stimulation (rTMS) treatment of upper-extremity motor dysfunction after stroke. Studies have shown that combined low- and high-frequency stimulation (LF-HF-rTMS) of the bilateral cerebral hemispheres is more effective than sham stimulation or stimulation of one cerebral hemisphere alone in treating motor dysfunction in the subacute stage of stroke. The efficacy of this protocol in the convalescence phase of stroke has rarely been reported, and its mechanism of action has not been clarified. In this study, we designed a prospective, single-blind, randomized controlled trial to investigate the efficacy and safety of different stimulation regimens for the treatment of upper extremity motor disorders in patients with convalescent stage stroke and aimed to explore the underlying mechanisms based on biomarkers such as brain-derived neurotrophic factor (BDNF). METHODS: Seventy-six subjects will be randomly divided into combined, low-frequency, high-frequency, and control groups based on the proportion of 1:1:1:1, with 19 cases in each group. All groups will have conventional rehabilitation, on top of which the combined group will receive 1 Hz rTMS in the unaffected hemisphere and 10 Hz rTMS in the affected hemisphere. The low-frequency group will be administered 1 Hz rTMS in the unaffected hemisphere and sham stimulation in the contralateral hemisphere. The high-frequency group will be administered 10 Hz rTMS in the affected hemisphere and contralateral sham stimulation. The control group will receive bilateral sham stimulation. Assessments will be performed at baseline, after 2 weeks of treatment, and at post-treatment follow-up at week 6. The primary outcomes are FMA-UE (Fugl-Meyer assessment-upper extremity), latency, and serum BDNF levels. The secondary outcomes are the National Institute of Health Stroke Scale (NIHSS), Brunnstrom staging (BS), modified Ashworth scale (MAS), Modified Barthel Index (MBI), central motor conduction time (CMCT), precursor proteins of mature BDNF (proBDNF), and matrix metalloproteinase-9 (MMP-9) levels. Adverse events, such as headaches and seizures, will be recorded throughout the study. DISCUSSION: The findings of this study will help develop optimal stimulation protocols for motor recovery in stroke patients and identify biomarkers that respond to post-stroke motor rehabilitation, for better guidance of clinical treatment. TRIAL REGISTRATION: The study protocol was passed by the Medical Research Ethics Committee of the General Hospital of Ningxia Medical University on January 1, 2022 (no. KYLL-2021-1082). It was registered into the Chinese Clinical Trials Registry on May 22, 2022 (no. ChiCTR2200060201). This study is currently in progress.

Assessing Major Accident Risks to Support Land-Use Planning Using a Severity-Vulnerability Combination Method: A Case Study in Dagushan Peninsula, China.

Major accident risks posed by chemical hazards have raised major social concerns in today's China. Land-use planning has been adopted by many countries as one of the essential elements for accident prevention. This article aims at proposing a method to assess major accident risks to support land-use planning in the vicinity of chemical installations. This method is based on the definition of risk by the Accidental Risk Assessment Methodology for IndustrieS (ARAMIS) project and it is an expansion application of severity and vulnerability assessment tools. The severity and vulnerability indexes from the ARAMIS methodology are employed to assess both the severity and vulnerability levels, respectively. A risk matrix is devised to support risk ranking and compatibility checking. The method consists of four main steps and is presented in geographical information-system-based maps. As an illustration, the proposed method is applied in Dagushan Peninsula, China. The case study indicated that the method could not only aid risk regulations on existing land-use planning, but also support future land-use planning by offering alternatives or influencing the plans at the development stage, and thus further enhance the roles and influence of land-use planning in the accident prevention activities in China.

[Expressions and significances of CD147, OPN and MMP-2 in oral squamous cell carcinoma].

OBJECTIVE: To evaluate the expressions and significances of CD147, OPN and MMP-2 in oral squamous cell carcinoma (OSCC). METHODS: The expressions of CD147, OPN and MMP-2 were detected by immunohistochemical method (SP) in 35 cases of ()SCC. The relationships between the expressions of CD147, OPN and MMP-2 and clinic histopathologic parameters of OSCC were analyzed. RESULTS: The expression rates of CD147, OPN and MMP-2 in OSCC tissues were 65.71%, 71.43% and 68.57% respectively. The expressions of CD147, OPN and MMP-2 were positively correlated with tumor histopathologic type, clinical stage and lymph node metastasis (P < 0.05), but not correlated with age, gender and the location of tumor (P > 0.05). The positive correlations were found among the expressions of CD147, OPN and MMP-2 (MMP-2 and CD147, r = 0. 653; MMP-2 and OPN, r = 0.540; CD147 and OPN, r = 0.381; P < 0.05). CONCLUSION: The expressions of CD147, OPN and MMP-2 increase in OSCC and may be the potential predictor of the malignant degree of OSCC.