In general, you shouldn't use dbfs any more, instead, use Volumes. 

But, as example, if I have an excel in my Worskpace directory, you could do this.

%pip install openpyxl
import pandas as pd

# replace with your path
file_path = "/Workspace/Users/stefan.koch@btelligent.com/excel/FinancialsSampleData.xlsx"

# read the sheet with Name Financials1 into a pandas dataframe
pdf = pd.read_excel(file_path, sheet_name="Financials1")

# Transform the Pandas Dataframe to a Pyspark Dataframe
df = spark.createDataFrame(pdf)

display(df)

Would this work for you or what is your dbfs path?